Heterogeneous receptor expression underlies non-uniform peptidergic modulation of olfaction in Drosophila

Sizemore, Tyler R.; Jonaitis, Julius; Dacks, Andrew M.

doi:10.1038/s41467-023-41012-3

Download PDF

Article
Open access
Published: 30 August 2023

Heterogeneous receptor expression underlies non-uniform peptidergic modulation of olfaction in Drosophila

Nature Communications volume 14, Article number: 5280 (2023) Cite this article

2268 Accesses
11 Altmetric
Metrics details

Subjects

Abstract

Sensory systems are dynamically adjusted according to the animal’s ongoing needs by neuromodulators, such as neuropeptides. Neuropeptides are often widely-distributed throughout sensory networks, but it is unclear whether such neuropeptides uniformly modulate network activity. Here, we leverage the Drosophila antennal lobe (AL) to resolve whether myoinhibitory peptide (MIP) uniformly modulates AL processing. Despite being uniformly distributed across the AL, MIP decreases olfactory input to some glomeruli, while increasing olfactory input to other glomeruli. We reveal that a heterogeneous ensemble of local interneurons (LNs) are the sole source of AL MIP, and show that differential expression of the inhibitory MIP receptor across glomeruli allows MIP to act on distinct intraglomerular substrates. Our findings demonstrate how even a seemingly simple case of modulation can have complex consequences on network processing by acting non-uniformly within different components of the overall network.

Long-range GABAergic projections contribute to cortical feedback control of sensory processing

Article Open access 12 November 2022

Target specific functions of EPL interneurons in olfactory circuits

Article Open access 29 July 2019

Morphology and physiology of olfactory neurons in the lateral protocerebrum of the silkmoth Bombyx mori

Article Open access 12 November 2019

Introduction

Animals use their sensory systems to internalize and process information about the identity, intensity, and valence of external stimuli, so they can properly navigate their environment. However, constant ecological and internal state fluctuations threaten the animal’s ability to accurately represent these stimuli. To address the demands these fluctuations impose, sensory systems use processes such as neuromodulation to flexibly adjust sensory processing and behavior. The largest and most ancient collection of neuromodulators are small peptides (~3–100 amino acids) termed neuropeptides^{1,2,3,4,5,6,7,8,9,10}. For instance, neuropeptide F (NPF)/neuropeptide Y (NPY) play a conserved role in promoting feeding behaviors in sea slugs, humans, flies, zebrafish, nematodes, mosquitoes, and rodents^{11,12,13,14,15,16,17,18,19}. Yet, despite their ubiquity and clear importance in nervous systems^20,21,22, the mechanistic basis of peptidergic modulation of sensory processing remains unclear.

Often peptidergic modulation is strongly associated with a given physiological drive, and in some cases the actions of a neuropeptide within a single network can be associated with different behavioral contexts. For instance, myoinhibitory peptide (MIP) is necessary and sufficient to stimulate the drive of Drosophila towards food-odors within the context of satiation²³, and is implicated in the post-mating shift in odor-preferences²⁴. However, MIP appears to be uniformly distributed across all antennal lobe (AL) glomeruli, which process far more than just food-related odors. Therefore, how can a ubiquitously distributed neuropeptide have stimulus-specific effects on sensory processing? Here, we reveal the cellular, physiological, and structural substrates that enable MIP to differentially modulate olfactory input to distinct olfactory channels.

Results

MIP differentially modulates olfactory input to distinct glomeruli

To test whether odorant responses within different glomeruli are uniformly modulated by MIP, we chose an odor which activates several glomeruli visible at the same/nearly the same imaging depth, and whose pattern of glomerular activation is well-established—apple cider vinegar (ACV)²⁵. In this way, any non-uniform effects of MIP on OSN odor-responses across different glomeruli would be readily detectable. Furthermore, we recorded the odor-evoked responses of OSN axon terminals (i) before MIP was pressure injected onto the AL (“pre-MIP injection”), (ii) after MIP was injected, but before it was removed from the perfusate (“MIP”), and (iii) after a brief washout period (“post-washout”) (see Methods) (Fig. 1a).

**Fig. 1: Myoinhibitory peptide (MIP) differentially modulates OSN ACV responses.**

Before MIP application, OSNs robustly respond to both test concentrations of ACV (Fig. 1b, c), then after MIP is pressure injected into the AL DM1 OSN responses to 10⁻² ACV are increased (Fig. 1b). Similarly, DM4 OSN responses to 10⁻⁶ ACV are also increased after peptide application (Fig. 1c). After a brief washout period, DM1 OSN responses to 10⁻² ACV return to pre-peptide application responses (Fig. 1b), whereas the increased DM4 OSN responses to 10⁻⁶ ACV are sustained (Fig. 1c). In contrast, DM2 OSN responses are substantially diminished after peptide application regardless of odor concentration, and remain so post-washout (Fig. 1b, c). Moreover, DM5 OSN responses to 10⁻² ACV, which were decreased (albeit insignificantly) upon peptide application, become significantly diminished post-washout relative to pre-peptide application (Fig. 1b).

Altogether, these results show that MIP can differentially modulate OSN odor-evoked responses in a glomerulus- and stimulus concentration-dependent manner, while also having concentration-independent consequences on OSNs of another glomerulus. However, these observations could be explained by differences in MIP-SPR signaling substrates across these glomeruli. For instance, there may be no synaptic input to DM1 OSNs from MIP-immunoreactive (MIP-ir) AL neurons, and therefore our prior observations (Fig. 1b, c) are the result of polysynaptic influences induced by MIP application. Therefore, we sought to test our suppositions by resolving the entire MIPergic signaling circuit architecture, including the identity of the presynaptic MIP-releasing AL neurons, their pre- & postsynaptic partners, and those postsynaptic partners that express SPR.

Patchy GABAergic LNs are the sole source of MIP within the Drosophila AL

Previous neuroanatomical investigations suggested that the neurites of the AL-associated MIPergic neurons appear restricted to the AL, which implies MIP is released from AL LNs²⁶. However, the Drosophila AL houses ~200 LNs whose distinct roles in olfactory processing have been associated with their transmitter content and morphology^{27,28,29,30,31,32,33,34,35,36,37}. For example, individual cholinergic AL LNs innervate many glomeruli and perform lateral excitation as a means for broadening odor representations in the AL^{33,34,38,39,40}. Therefore, to resolve whether MIP-ir AL neurons are indeed AL LNs, and if they belong to a known AL LN chemical class, we assessed the overlap of MIP-immunoreactivity with markers for the major Drosophila small-neurotransmitters⁴¹ (Fig. 2a and Supplementary Fig. 1). We find that AL MIP-ir neurons do not overlap with choline acetyltransferase (ChAT) or vesicular glutamate transporter (VGlut), but all MIP-ir neurons in the AL overlap with GAD1 (9.1 ± 0.19 neurons, n = 5 brains, 10 ALs) (Fig. 2a and Supplementary Fig. 1). In accordance with RNA-sequencing^42,43,44, we find no detectable MIP-immunoreactive OSNs (Supplementary Fig. 1). Altogether, these results suggest that an ensemble of ~9 GABAergic LNs are the source of MIP within the Drosophila AL.

**Fig. 2: Myoinhibitory peptide (MIP) is released by GABAergic patchy LNs in the AL.**

The Drosophila AL houses a variety of distinct GABAergic LNs, which can be subdivided into five major morphological types: panglomerular, multiglomerular, oligoglomerular, continuous, and patchy³². Like cortical interneurons^45,46, these different interneuron morphological types play distinct roles in AL olfactory processing. To determine morphological type to which the MIPergic LNs belong, we screened the Janelia FlyLight driver line collection⁴⁷, tested ~25 of those lines for MIP-immunoreactivity, and identified a GAL4 driver (R32F10-GAL4) that selectively highlights MIPergic LNs within the AL (Fig. 2b–d). We then combined R32F10-GAL4 with a GFP-tagged rat preproatrial natriuretic factor (ANF-GFP) which, when expressed within peptidergic neurons, is proteolytically processed and packaged into secretory vesicles and preferentially accumulates in peptidergic synaptic terminals⁴⁸. We find broad ANF-GFP accumulation in R32F10-GAL4 AL LN terminals across the AL, confirming these LNs possess the necessary subcellular machinery for neuropeptide processing, packaging, and release (Fig. 2e). Furthermore, all MIP-ir is abolished in the AL when R32F10-GAL4 AL LNs are ablated via temperature-gated expression of a cell-specific diphtheria toxin (Fig. 2f). We then leveraged our selective genetic access to these LNs to resolve the morphology of individual MIPergic LNs through stochastic labeling⁴⁹. From these experiments, we find that all MIPergic LNs have a discontinuous innervation pattern resembling that of patchy AL LNs (Fig. 2g and Supplementary Fig. 1). Taken together, our data suggests that ~9 GABAergic patchy LNs are the sole source of MIP within the Drosophila AL.

There are many AL neurons, including other LNs, that are not patchy LNs but resemble the discontinuous morphology described for patchy LNs³². However, individual patchy LNs are unique in that they innervate different sets of glomeruli from animal-to-animal³². Therefore, we analyzed the set of glomeruli innervated by 50 individual MIPergic LNs and observed 50 distinct innervation patterns, thus demonstrating that no individual MIPergic LN innervates the same set of glomeruli across animals (Fig. 2h and Supplementary Figs. 2 and 3). Additionally, we find individual MIPergic LNs do not preferentially innervate any one glomerulus over others (Supplementary Fig. 2). When sister clones were assessed, we find that two individual MIPergic LNs co-innervate ~12 glomeruli (n = 5 brains, 5 sister clones) (Supplementary Fig. 3a, b), and individual MIPergic LNs consistently innervated at least one of the hygro-/thermosensory associated glomeruli (Supplementary Fig. 3c–e). These results suggest that at least two MIPergic LNs innervate any single glomerulus, including hygro-/thermosensory glomeruli^50,51,52. Moreover, these observations also demonstrate that individual MIPergic LNs innervate different glomeruli from animal-to-animal.

Most odorants activate more than one glomerulus in the Drosophila AL^25,53,54,55. Thus, if individual MIPergic LNs innervate different sets of glomeruli from animal-to-animal, are there pairs of glomeruli that are innervated significantly more than others? If so, what ecological relationships exist amongst significantly correlated pairs of glomeruli? To determine the probability that an individual MIPergic LN that innervates one glomerulus will innervate/avoid another glomerulus, we leveraged our previous clonal analysis data (Fig. 2g, h) to calculate a correlation coefficient for all possible pairs of glomeruli (Fig. 2i). This analysis revealed several statistically significant relationships, of which the most significant pairs were DM3-D (r = 0.49, p = 2.7 × 10⁻⁴) and VL2p-VA6 (r = −0.47, p = 4.9 × 10⁻⁴) (Fig. 2i). In addition to DM3-D and VL2p-VA6, this analysis also revealed a significant probability for MIPergic LN co-innervation amongst several pairs of glomeruli responsive to ACV²⁵, such as VM2-DM1 (r = 0.35, p = 0.01), DM4-DM2 (r = 0.31, p = 0.03), and DP1m-DM1 (r = 0.29, p = 0.04). This suggests that the glomerular innervation patterns of individual MIPergic LN likely do not explain non-uniform MIPergic modulation of OSN (Fig. 1). However, it is plausible that differential modulation of OSN responses by MIP arises as a consequence of the non-uniform MIPergic LN pre-/postsynaptic sites across these glomeruli. Therefore, we sought to determine whether MIPergic LN input/output sites were heterogeneously distributed to any particular glomeruli throughout the AL.

MIPergic LNs provide and receive broad input and output across the AL

We have shown that no individual MIPergic LN innervates the same set of glomeruli from animal-to-animal (Fig. 3a), but every glomerulus is innervated by at least one MIPergic LN across all animals (Fig. 3b). Therefore, we wondered whether significant differences in MIPergic LN input/output between glomeruli exist (Fig. 3c), which would explain the non-uniform effects of MIPergic modulation. To test this, we measured the density of MIPergic LN-expressed mCD8::GFP, anti-MIP immunoreactive punctae, and the synaptic polarity markers DenMark and synaptotagmin.eGFP (syt.eGFP)^56,57 in each glomerulus across many animals (Fig. 3d, e). We find that the density of each indicator varies across glomeruli but are stereotypic across samples (Fig. 3e and Supplementary Fig. 3f–i). The density of the output indicators (syt.eGFP and MIP-ir puncta) were statistically correlated, and nearly every indicator scaled with MIPergic LN intraglomerular cable density (Supplementary Fig. 3f–i). Even so, we find within-indicator voxel densities are generally evenly distributed across each glomerulus, suggesting MIPergic LN input and output are evenly distributed across the AL (Fig. 3e).

**Fig. 3: MIPergic LN input and output sites throughout the entire AL.**

These puncta analyses afford the advantages of analyzing MIPergic LN synaptic polarity across many individuals of both sexes. However, light microscopy is limited by its inability to resolve fine structures such as axons/dendrites^58,59. Therefore, we performed similar analyses on individual putative MIPergic LNs (putMIP LNs) within the most densely-reconstructed Drosophila central brain EM volume to-date, the hemibrain^60,61. Additionally, EM-level analyses of putMIP LN connectivity have the added benefit of shedding light on what type(s) of neuron(s) and/or stimuli might generally promote MIP recruitment in AL processing. More specifically, EM analysis of putMIP LN connectivity allowed us to determine: (i) do certain glomeruli receive more input from putMIP LNs (and vice versa) than others? (ii) what neurons are upstream/downstream of putMIP LNs in each glomerulus? and, (iii) at which putMIP LN presynaptic terminals are vesicles associated with neuropeptides (dense core vesicles, or DCVs)^62,63 found? Thus, we first used several criteria to identify fully-reconstructed putMIP LNs, such as the candidates’ principal identity, previous AL LN subtyping results²⁸, synaptic connectivity, and neuromorphic similarity to R32F10-GAL4 AL LNs (see Methods) (Fig. 3f). These stringent criteria resulted in the identification of 14 ideal candidates (Fig. 3f).

After identifying several optimal candidates, we tested whether any putMIP LNs have distinct dendritic and axonic compartments. If true, this would suggest putMIP LNs make region-specific input/output, as has been suggested for the “heterogeneous LNs” in the honeybee AL^64,65,66,67. Synaptic flow centrality and axonal-dendritic segregation indices⁶⁸ reveal all putMIP LNs lack clearly separable input/output compartments (Fig. 3g). When we assess the ratio of input-to-output along a given putMIP LN’s intraglomerular neurites, we find that the amount of input a given putMIP LN receives typically outnumbers the amount of putMIP LN output within any given glomerulus (Fig. 3h). To better understand how these inputs might drive MIPergic modulation, we assessed the general identity of all inputs a putMIP LN receives, as well as the class and transmitter type of each presynaptic input an intraglomerular putMIP LN arbor receives.

Generally, nearly half of all putMIP LNs receive more input from other LNs than other principal neuron categories (6/14 putMIP LNs; ~38-40% total input) (Fig. 4a). Similarly, just as many putMIP LNs receive the majority of their input from PNs than any other principal neuron category (6/14 putMIP LNs; ~31–33% total input) (Fig. 4a). Additionally, we found that within a given glomerulus putMIP LNs largely avoid one another, but do occasionally form synaptic connections (Fig. 4b and Supplementary Fig. 3j, k). When we refine these analyses by considering the putMIP LN’s intraglomerular connectivity and the presynaptic partner’s identity, we find the amount of excitatory or inhibitory input a given putMIP LN receives varies greatly across glomeruli. However, in every case, putMIP LNs generally receive far more excitatory than inhibitory input within any given glomerulus (~65–93% of all glomeruli innervated by the given putMIP LN) (Fig. 4b). This suggests that MIPergic LN intraglomerular processes may be broadly activated by disparate odorants, which would suggest uniform release of MIP may occur in response to various odors. Therefore, we tested whether MIPergic LNs are broadly activated in vivo by chemically diverse odorants.

**Fig. 4: Anatomical inputs to putMIP LNs and functional glomerular outputs from identified MIPergic LNs.**

MIPergic LNs generally display glomerulus-specific odor-evoked responses

Synapse counts have been shown to strongly predict functional output strength in neurons within other systems, including other Drosophila AL neurons^{28,69,70,71,72,73,74}. Our connectomic analyses intraglomerular putMIP LN arbors generally receive mostly excitatory input within a given glomerulus (Fig. 4b), which would imply that MIPergic LNs are broadly activated regardless of odor identity. This would be consistent with previous characterizations of large ensembles of GABAergic AL LNs odor-evoked GCaMP responses that have observed odor-invariant activation across nearly all glomeruli^54,75. Therefore, we recorded the in vivo odor-evoked responses of MIPergic LN intraglomerular neurites to a panel of chemically diverse odorants across multiple glomeruli (Fig. 5a). Moreover, we chose to image within glomeruli whose cognate PNs respond to at least one odorant in our test panel to better understand what role the MIPergic LNs may play in the OSN-to-PN information transfer. For example, benzaldehyde and geranyl acetate each evoke responses in DM1, DM4, DP1l, VA2, and VM2 PNs^37,76,77,78.

**Fig. 5: MIPergic LNs are differentially activated by different odors.**

In contrast to other GABAergic AL LNs^54,75, we find MIPergic LNs generally display glomerulus-specific responses to all test odors (Fig. 5a). For example, MIPergic LN neurites within VM2 and DP1m—two glomeruli visible at the same imaging depth—are simultaneously activated and inhibited by 1-octen-3-ol, respectively (Fig. 5a–c). This same odor drove post-excitatory depression in MIPergic LN neurites within DP1l and VA2 (Fig. 5a–c). However, in many cases MIPergic LN intraglomerular neurites did not respond to the given odorant (Fig. 5a–c). This shows that, unlike other GABAergic AL LNs which are broadly activated in response to similar stimuli^33,54,75, MIPergic LN intraglomerular processes are differentially activated by different odors. However, we acknowledge that these odor-evoked responses do not necessarily reflect MIP release itself, which we were unable to test for reasons described below (see Discussion).

Notably, ACV elicited robust activation of MIPergic LN intraglomerular processes across all identifiable glomeruli (Fig. 5a–c), including those glomeruli we found MIP non-uniformly modulates (Fig. 1). This finding, together with our earlier results (Figs. 2–5), suggests that the non-uniform effects of MIP on olfactory input likely do not arise from the presynaptic MIP-releasing neurons themselves. Instead, these results suggest the non-uniform effects of MIP on olfactory input are an emergent property of either (i) MIPergic LN postsynaptic targets, and/or (ii) differential SPR expression across the AL.

MIPergic LN downstream partners and widespread SPR expression within the AL

To determine the AL principal neurons likely targeted by MIPergic LNs, we first assessed the general output demographics for each putMIP LN (Fig. 6a, b). Of the AL principal neuron types, most putMIP LNs chiefly target PNs (71% of putMIP LNs; ~19–31% of putMIP LN total output) (Fig. 6a). The remaining minority of putMIP LNs are chiefly presynaptic to OSNs (29% of putMIP LNs; ~27–33% of putMIP LN total output) (Fig. 6a). Since AL PNs express GABA_A and GABA_B receptors²⁷, these results would suggest that MIPergic LNs may provide fast- and slow-acting inhibition across the AL, perhaps as a means to normalize PN odor-evoked responses. To determine which of these downstream partners (Fig. 6b) are likely targeted by MIPergic modulation, we determined which postsynaptic partners were downstream of putMIP LN terminals where dense core vesicles (DCVs) are observable (Fig. 6c). We observed several instances where DCVs could be found in putMIP LN terminals presynaptic to OSNs, PNs, and ventral LNs (Fig. 6c). However, MIPergic LNs could also release other neuropeptides, so the presence of DCVs in MIPergic LN presynaptic terminals does not necessarily mean the downstream neuron is modulated by MIP. Moreover, putMIP LN EM analyses indicate several AL principal neuron types are plausible targets for MIPergic modulation (Figs. 3, 6, and 7a). To determine which downstream partners are subject to MIPergic modulation, we identified the AL neurons that express MIP’s cognate receptor, the G_α-i/G_α-o-coupled sex peptide receptor (SPR)^79,80,81,82. To do so, we used a CRISPR/Cas9-mediated T2A-GAL4 insertion within the endogenous SPR locus to enable GAL4 expression within SPR-expressing cells⁸³ (Fig. 7b).

**Fig. 6: Postsynaptic targets of each putMIP LN and representative putMIP LN presynaptic terminals with dense core vesicles (DCVs).**

**Fig. 7: Widespread sex peptide receptor (SPR) expression throughout the AL.**

In Drosophila, OSN somata are located within the third-antennal segment and maxillary palp^84,85. We find 208.9 ± 11.89 (n = 17 animals, 30 antennae) and 63.42 ± 4.31 (n = 18 animals, 31 maxillary palps) SPR-T2A-GAL4⁺ neurons in the third-antennal segment and the maxillary palp, respectively (Fig. 7c, d). As there are ~945 and ~113 OSNs in the antennae and maxillary palps, respectively⁸⁶, this would suggest that ~22% of antennal OSNs and ~56% of maxillary palp OSNs express SPR. The number of SPR-T2A-GAL4⁺ neurons in either appendage do not significantly differ based on the animal’s sex or mating status (antennae: p = 0.107, one-way ANOVA; maxillary palps: p = 0.559, Kruskal-Wallis test). However, this does not discount differences in the level of SPR expression within these neurons based on the animal’s sex or mating status. Through stochastic labeling experiments where the antennal nerve is left attached to the brain, we found OSN fibers that innervate many distinct glomeruli, including several ACV-responsive OSNs (Fig. 7e and Supplementary Data 1). Interestingly, we found SPR-T2A-GAL4 expression in afferents belonging to every sensory modality (Supplementary Fig. 4), which suggests MIPergic modulation of sensory afferents may be a fundamental feature in Drosophila.

Within the brain, we find consistent colocalization of SPR-T2A-GAL4 and the proneural gene embryonic lethal abnormal vision (ELAV) (Fig. 7f) and the glial marker reverse polarity (REPO) (Supplementary Fig. 4). Through intersectional genetics and stochastic labeling, we find that these ELAV-positive neurons are composed of: 4.89 ± 0.21 (n = 23 brains, 44 ALs) SPR-expressing ventral glutamatergic LNs (GlutLNs) (Fig. 7f, g), uniglomerular PNs (Fig. 7h and Supplementary Data 1), and several lateral LNs (Fig. 7i). In agreement with these results, we find similar neuron types using another SPR driver (SPR-GAL4::VP16)⁸⁷ (Supplementary Fig. 5), several publicly available scRNA-seq datasets^88,89,90 (Supplementary Fig. 6), and a novel SPR^MI13553-T2A-LexA::QFAD driver (Supplementary Fig. 7). Notably, SPR-T2A-GAL4⁺ neurons were consistent across all animals—a stark contrast to the animal-to-animal differences in individual MIPergic LN morphology previously observed (Fig. 2).

Differential SPR expression across glomeruli enables non-uniform MIPergic modulation of olfactory input

To test the necessity of direct MIP-SPR signaling on modulation of OSN odor-evoked responses, we repeated our earlier experiments (see Fig. 1), but used RNA interference (RNAi) to knockdown SPR specifically within OSNs (Fig. 8a, b). Moreover, this SPR-RNAi has been used to effectively knockdown SPR in OSNs previously²⁴, and abolishes SPR immunoreactivity in the Drosophila CNS when expressed pan-neuronally⁹¹. We find that SPR knockdown abolishes the MIP-induced decrease in the odor-evoked responses of DM2 and DM5 OSNs (Fig. 8c, d). This result is consistent with SPR-expression in DM2 and DM5 OSNs (Fig. 7e), and suggests MIP directly decreases the odor-evoked responses of DM2 and DM5 OSNs. In contrast, SPR knockdown in DM1 and DM4 OSNs does not prevent their responses from increasing after peptide application (Fig. 8c, d). Since we did not observe SPR-expression in DM1 and DM4 OSNs (Fig. 7), and SPR knockdown in these OSNs does not abolish the MIP-induced increase in their responses (Fig. 8c, d), our results suggest MIP acts polysynaptically to disinhibit (thus, increasing) DM1 and DM4 OSN odor-evoked responses (Fig. 8e).

**Fig. 8: SPR knockdown in OSNs reveals heterogeneous SPR expression across glomeruli enables non-uniform MIPergic modulation of OSN ACV responses.**

Discussion

Our data reveals the circuit topology that enables a single neuropeptide, acting through a single receptor, to differentially modulate olfactory processing. We show that pharmacological application of MIP elicits non-uniform and complex effects on olfactory input to the Drosophila primary olfactory center. Here, MIP reduces the responses of OSNs in some glomeruli, and simultaneously enhances the responses of OSNs in other glomeruli (Fig. 1). We show that the non-uniform effects of MIP on olfactory input is likely not an emergent property of the identity, structure, and/or connectivity of the MIP-releasing neurons, themselves. Instead, we find that differential SPR expression within distinct glomeruli enables MIP to non-uniformly modulate olfactory input across olfactory channels.

We found that individual MIPergic LNs innervate a different repertoire of glomeruli across animals and do not preferentially innervate any one glomerulus over others (Fig. 2 and Supplementary Fig. 2). These findings are consistent with earlier reports wherein patchy AL LNs were first generally described³². But, what factor(s) give rise to the tremendous flexibility within this single morphological subtype? One explanation might be that MIPergic LN morphological idiosyncrasy is a byproduct of experience during development. However, OSN removal in the adult does not disrupt the animal-to-animal variability of patchy LNs³². To the best of our knowledge, a single locus (e.g., environmental experience or heritable trait) that would support animal-to-animal variation in patchy LNs has not been identified.

Another explanation for animal-to-animal differences in individual MIPergic LN morphology is that it may not matter which individual MIPergic LN forms synapses with which downstream target, as long as all of the MIPergic LN downstream targets are met. Every nervous system is the byproduct of the adaptive pressures demanded by the animal’s niche; a place that can continually change in seemingly unpredictable ways. Therefore, a developmental parameter space may exist, wherein just enough genetic idiosyncrasy is allowed to help prevent extinction in the face of environmental perturbations. The breadth of this developmental parameter space (or the degree of variability from the median) would be defined by many generations of selective pressures, wherein subtle changes in genetic idiosyncrasies might equally result in winners and losers. As a consequence of these genetic idiosyncrasies, phenotypic variability in a given developmental program would inevitably accumulate, resulting in the observed animal-to-animal variability in neuronal features (e.g., morphology, ion channel distribution, etc.). Consistent with this idea, animal-to-animal variations in neural circuitry have been noted in grasshoppers⁹², crabs^{93,94,95,96,97}, lobsters^98,99, flies^32,100,101, and rats¹⁰². Moreover, inter-animal variations in neuronal architecture are one of several features implicated in inter-animal behavioral variations^{101,103,104,105,106,107}. However, despite this variability, overall neuronal circuit functions persist including consistent MIPergic LN synaptic polarity marker density (Fig. 3), MIPergic LN within-odor responses (Fig. 5), MIP-induced decreases in DM2 and DM5 OSN responses across animals (Fig. 6), and SPR expression (Fig. 7). Moreover, several positive and negative correlations exist for pairs of glomeruli innervated by single MIPergic LNs, such as the significant probability for MIPergic LN co-innervation in ACV-responsive glomeruli (Fig. 2i). Together, these results suggest that the morphology of an individual MIPergic LNs can differ from animal-to-animal, as long as the right combinations of downstream targets (e.g., ACV-responsive neurons) are met by the ensemble.

We have shown that a small ensemble (~5% of all AL LNs) of GABAergic AL LNs are the sole source of MIP in the Drosophila AL (Fig. 2 and Supplementary Fig. 1). This implies that MIPergic LNs have the capacity to adjust AL olfactory processing through both GABA- and MIP-release. Previous works found that the iono- and metabotropic GABA receptors are expressed amongst all AL principal neuron types^27,29,108, and we show SPR is similarly expressed by members of every AL principal neuron type (Fig. 7). Therefore, MIPergic LN activation could plausibly cause both fast-acting and long-lasting inhibition in the same and/or disparate downstream target. Moreover, MIPergic LN-derived GABA and MIP may simultaneously act on the same downstream target(s) to synergistically modulate their activity to have a greater effect than either modulator alone might achieve. Alternatively, MIPergic LNs might primarily use GABA throughout the course of ongoing network activity, and use MIP only under special circumstances (see example below).

We attempted to parse the contribution of GABA and MIP over the course of our investigation. We conceived an experimental paradigm in which MIPergic LNs would be artificially activated chemogenetically via P2X2 misexpression¹⁰⁹ and ATP injection, while we recorded OSN GCaMP responses before and after MIPergic LN activation. The results of these experiments would then be compared to similar experiments wherein MIPergic LN GAD1 expression would be knocked down as a means of resolving the contribution of MIPergic LN-derived GABA. However, both approaches require prior knowledge of what level of artificial activation is necessary to mobilize MIPergic LN dense core vesicles (DCVs), and thus MIPergic LN-derived MIP release.

To resolve the minimal strength of artificial activation necessary to mobilize MIPergic LN DCVs, we artificially activated MIPergic LNs (described above), while simultaneously recording DCV changes via either ANF-GFP⁴⁸ or the neuropeptide release sensor NPPR-ANP-GCaMP6s¹¹⁰. However, we were unable to detect any change in either indicator even when we injected 100 mM ATP—a concentration 10x-greater than what is necessary to activate other AL LNs¹¹¹. As a consequence, it remains infeasible to simply artificially activate MIPergic LNs, while measuring a downstream neurons’ responses, and accurately attribute changes in the downstream neuron MIP released from MIPergic LNs.

Co-transmitters often increase the computational capacity of the neuron they are released from, and the plasticity of the networks in which they act^{112,113,114,115}. Therefore, it is likely that future investigations using novel technology may find that the MIPergic LNs play a far greater role in AL processing than what is resolved here. As detailed above, each co-transmitter may confer unique and substantial contributions to the overall role these interneurons have in AL processing. As such, this aspect of MIPergic LN function is an important and remaining question that should be addressed by future investigations.

Generally, multiple glomeruli are activated by any given odorant^{25,53,54,55,116}. However, optogenetics can be used to selectively activate individual glomeruli in a manner similar to their odor-evoked responses to evaluate the behavioral contribution of individual glomeruli¹¹⁷. Such experiments reveal that DM1 and DM2 co-activation do not summate, and co-stimulation of both glomeruli produces a behavioral response that resembles DM1-only activation¹¹⁷. Based on this, the existence of an antagonistic relationship between DM1 and DM2 was proposed, wherein co-stimulation reduces the efficacy of either or both glomeruli¹¹⁷. We find MIP indirectly increases DM1 and directly decreases DM2 OSN responses (Fig. 8). Therefore, MIP-SPR signaling in DM1 and DM2 may act as a homeostat such that coactivation of each glomerulus never produces a behavioral response greater than the DM1-only activation response. This buffer would be advantageous for preventing saturation at the downstream neurons that receives convergent input from these glomeruli^{28,70,118,119}.

MIP-SPR signaling has been implicated in several behavioral state switches^23,24. Notably, abolishing MIP release by inactivating all MIPergic neurons, or using a MIP-genetic null mutation, increases the animal’s drive for food-derived odors²³. Moreover, DM2 OSN firing rate increases when all MIPergic neurons are inactivated²³. In contrast, increasing the activity of all MIPergic neurons decreases attraction toward food-odors, to the extent of eliciting odor-induced aversion²³. Together, these behavioral results suggest MIP-SPR signaling can affect the sensitivity to food-associated odors and drive to search for food. In accordance with these observations, we found that individual MIPergic LNs significantly co-innervate several food-odor associated glomeruli (Fig. 2) and neurons from several of these glomeruli express SPR (Fig. 7). Most strikingly, we find that MIP directly acts on DM2 OSNs to decrease their odor-evoked responses (Fig. 8). Furthermore, we show that the MIP-induced decrease in DM2 responses occurs in a stimulus-concentration independent manner (Figs. 1 and 8). Altogether, these results point to a probable role for MIPergic LN-derived MIP signaling to adjust olfactory processing, likely while other MIPergic neurons adjust other sensory/motor elements, in accordance with satiety homeostasis drives. However, this role is likely only one of many that the MIPergic LNs play in AL processing as they also release GABA, and form reciprocal connections with neurons outside of the SPR-expressing neurons (Figs. 3 and 4).

Peptidergic modulation can be as simple as a single neuropeptide modulating motor output in the stick insect locomotor system¹²⁰, or as complex as the ~37 neuropeptide families acting within the cortex¹²¹. Our data highlight how even a seemingly simple case, a single neuropeptide acting through a single receptor, can have complex consequences on network processing by acting non-uniformly within different components of the overall network. As neuropeptide functions are often deeply conserved, and as the actions of neuropeptides begin to come into focus, similar instances of complex and non-uniform peptidergic modulation will likely appear across disparate taxa and modalities.

Methods

Fly husbandry, genotypes, and subject details

A complete table of each animal’s genotype used for each experiment are included in Supplementary Data 1. Information on parental stock origins and relevant identifiers are provided in Table 1. Unless otherwise noted, flies were reared on standard cornmeal and molasses media at 24 °C and under a 12:12 light:dark cycle. Equal numbers of male and female animals were used when possible, excluding live-imaging experiments which used only females. For mating status comparisons: 1) virgin females denotes females that were meconium-positive upon collection, 2) non-virgin females were housed with males until processing for immunohistochemistry, and 3) flies were age-matched and kept on the similar media until processed for immunohistochemistry.

Table 1 Sources and identifiers for all key reagents and resources used in this present study

Full size table

Immunohistochemistry and imaging

All immunohistochemistry was performed generally as previously described¹²². Briefly, samples were dissected, fixed in 4% paraformaldehyde, then washed with phosphate buffered saline with 0.5% Triton-X 100 (PBST) several times before taking samples through an ascending-descending ethanol wash series, then blocking in 4% IgG-free BSA (Jackson Immunoresearch; Cat#001-000-162). Samples were then incubated in primary antibody (Table 1) diluted in blocking solution and 5 mM sodium azide. Following primary antibody incubation samples were washed with PBST, blocked, and incubated in secondary antibody diluted in blocking solution and 5 mM sodium azide. Finally, samples were washed, cleared using an ascending glycerol series (40%, 60%, 80%), and mounted on well slides in Vectashield® (Vector Laboratories, Burlingame, CA; Cat#H-1200). Images were collected and analyzed as previously described¹²² with VAA3D¹²³ and FluoRender¹²⁴, apart from those captured with a 40x/1.25 Silicone UPlanSApo Olympus objective. In some cases, image brightness and contrast were manually adjusted in CorelDRAW 2021 (Corel Corp., Ottawa, Canada). In the case of Fig. 7h, i, the FluoRender brush tool¹²⁴ was used to select and segment the relevant neuron(s) if they were obscured by the primary neurites of other cell(s) that occlude the cells-of-interest. All original light microscopy confocal scans used to create the figure panels are available at https://zenodo.org/10.5281/zenodo.8127341.

Single LN clone induction and glomerular innervation analyses

Single LN clones were induced through the MultiColor Flip Out (MCFO) method⁴⁹. Flies carrying the MCFO cassettes, Flp-recombinase, and GAL4 driver were raised under normal conditions (see above) until heat shock. Adult flies were heat-shocked in a 37 °C water bath for 12–25 min and returned to normal conditions for ~2–3 days before processing for immunohistochemistry. We chose to analyze the innervation patterns of 50 individual MIPergic LNs based on a statistical probability theorem termed, the coupon collector problem¹²⁵. For our purposes, this meant we needed to sample 43 individual LNs to ensure we sampled each of the ~13 LNs highlighted by R32F10-GAL4 (Fig. 1b–d). We chose to analyze more than the minimal number as determined by this theorem as an additional preemptive measure to ensure the ~8 MIPergic AL LNs were sampled. Apart from VA1v, glomeruli were defined according to previously published AL maps^126,127. Glomerulus names were later updated according to recent naming conventions²⁸. Neuropil were labeled using anti-DN-cadherin or anti-Bruchpilot (Table 1). Hierarchical clustering and principal components analysis (PCA) of glomerular innervation data were performed as previously described³². PCA was performed without any arbitrary threshold of significance. Glomeruli and individual MIPergic LN clones were hierarchically clustered using Ward’s method (“ward.D2”) and Euclidean distance using the “Heatmap” function in the ComplexHeatmap package¹²⁸. Pairwise Pearson’s correlation coefficient for all possible binary combinations of glomeruli were determined from our MIPergic LN glomerular innervation clonal analysis data using the “cor” function in the base-R stats package. These Pearson’s correlation coefficients were subsequently assessed for statistical significance by using the “rcorr” function in the Hmisc package, which computes a matrix of Pearson’s r rank correlation coefficients for all possible pairwise combinations within a data matrix. The p-values in this instance are the probability that we would have found a given result if the correlation coefficient was zero (the null hypothesis). This is an indication of whether the aforementioned co-innervation of a given pair of glomeruli is significant or not. In other words, if glomerulus A and B are likely co-innervated by a given MIPergic LN (i.e., positive correlation of MIPergic LN between glomerulus A and B), is this likelihood statistically probable? Further details regarding how significant correlations are computed using this approach are provided in the package’s documentation (https://github.com/harrelfe/Hmisc/blob/master/R/rcorr.s). The corrplot package was used to create the hierarchically clustered (using Ward’s method) representation of these pairwise correlation coefficients (Fig. 1). In every case used, glomerular “odor scene” information is derived from previous assignments¹²⁹.

To determine if MIPergic LNs preferentially innervate glomeruli based on valence, glomeruli were assigned “attractive” or “aversive” based on similar assignments previously described^28,129. These glomerular valences aggregate findings from previous reports^{25,130,131,132,133,134}, as well as behavioral valence of the odors¹³⁵ that glomerulus’ OSNs respond to according to DoOR 2.0¹³⁶. Glomeruli whose valence is state-dependent (e.g., the V glomerulus)¹³⁷ and DC4 were not included in this analysis. Similar methods were used to determine if MIPergic LNs preferentially innervate glomeruli based on the functional group of a given OR’s cognate odorant, with the exception of the V and VM6 glomeruli.

MIPergic LN ablation experiments

To determine whether MIPergic LNs are the sole source of MIP immunoreactivity within the AL, we used R32F10-GAL4 to drive the expression of a temperature-sensitive variant of Diphtheria toxin (UAS-DTI)¹³⁸ in all MIPergic LNs. Animals carrying both transgenes were raised at a permissive temperature of 18 °C, until ~2 days post-eclosion when they were shifted to the non-permissive temperature of ~25–28 °C for ~3 days. After ~3 days at 25 °C or 28 °C animals were processed for MIP-immunolabeling as described above.

MIPergic LN anatomical marker density analyses

Analysis of syt.eGFP, DenMark, anti-MIP immunoreactive puncta signal, and LN innervation density in antennal lobe glomeruli (via mCD8::GFP signal) was performed as previously described⁷⁵. Images of all antennal lobes within a given brain were collected with similar confocal scan settings (laser power, detector offset, etc.) and later imported into FIJI for quantification. Using the Segmentation Editor plugin and a previously described script (graciously provided by Rachel Wilson, Harvard)⁷⁵, ROIs were manually traced every 2–3 slices around the neuropil boundaries of each glomerulus using the anti-DN-Cadherin or anti-Bruchpilot channel, and then interpolated through the stack to obtain boundaries in adjacent slices. To ensure each brain contributed equally when pooling data across brains, signal density values for all glomeruli were normalized to the maximum density value within the given indicator being analyzed (e.g., all density values for syt.eGFP were normalized to the maximum syt.eGFP value). Signal density values were similarly normalized within-indicator, but also within-sex, for assessing sexual dimorphism in MIPergic LN syt.eGFP or DenMark puncta signal (Supplementary Fig. 7). The “ggscatter” function in the ggpubr package was used to determine Pearson’s correlation coefficients and p-values when assessing correlations between effector/anti-MIP and MIPergic LN mCD8::GFP voxel density across all glomeruli. Adjusted R² values were calculated using the base-R stats package and correspond to how well each data being assessed for the given correlation analysis fit a linear model.

Putative MIPergic LN connectomic analyses — identifying putative MIPergic LNs

All connectome analyses leveraged the publicly available Janelia FlyEM Drosophila hemibrain electron microscopy volume (v.1.2.1; https://neuprint.janelia.org/)^60,61, and recently described analysis suites^28,139. We used several stringent criteria for determining which neurons are most likely MIPergic LNs, the first of which was the candidate neurons must be AL LNs. We then selected those candidate AL LNs that were previously determined to most likely belong to the patchy AL LN subtype²⁸. Candidates were then filtered for those that receive input from the contralaterally projecting, serotonin-immunoreactive deutocerebral (CSD) neurons as all MIPergic LNs express the 5-HT1A serotonin receptor¹²², and form connections with the serotonergic CSD neurons¹⁴⁰. We then used natverse¹³⁹ to transform the interconnectivity of each candidate neuron into the FlyCircuit whole brain (FCWB) template brain three-dimensional space^141,142, so we could generate a morphological similarity score between our query neuron and neurons FlyLight project’s GMR-GAL4 repository⁴⁷ by using the built-in NBLAST package (nat.nblast)¹⁴². We selected for only those candidates that achieved a GMR32F10-GAL4 NBLAST score of >0.80, which is greater than the >0.60 score necessary to consider the query neuron and GMR GAL4 neurons identical twins¹⁴². Lastly, any remaining candidate MIPergic LNs were filtered for those neurons that are considered the hemibrain’s highest level of tracing completeness and confidence (“Traced”). Only neurons that met all of these criteria (~7% of all AL LNs) were considered for further analysis.

Despite these stringent criteria, these procedures resulted in the identification of 5 more neurons than there are MIP-immunoreactive neurons in the Drosophila AL (14 ideal candidates vs. ~9 MIP-immunoreactive neurons). This means that 5/14 putative MIP LNs analyzed here are likely not MIP AL LNs. As all known biological characteristics of MIPergic AL LNs were exhausted to identify these 14 ideal candidates, there is no convincingly reasonable method to identify which 5 putative MIP LNs should be excluded. Therefore, we elected to analyze all 14 ideal candidates (as opposed to imposing an arbitrary criteria) to ensure any/all MIPergic AL LN connectomic topological properties would be captured. That said, although ~36% of the putMIP LNs analyzed here are likely not MIPergic AL LNs, we generally find similar results for all 14 ideal candidates (Figs. 3, 4 and Supplementary Fig. 3). Thus, we do not find significant distinguishing characteristics of any 5 of these putative MIP LNs that suggests our reported interpretations are incorrectly attributable to their inclusion.

Putative MIPergic LN connectomic analyses — putMIP LN meshes, segregation indices, and flow centrality

Most methods for analyzing putMIP LN morphology and connectivity have been described recently²⁸. Putative MIPergic LN skeleton meshes (Fig. 3g) were fetched from the hemibrain data repository by accessing the neuPrint Python API using the neuprint-python (https://github.com/connectome-neuprint/neuprint-python) and Cloud-Volume (https://github.com/seung-lab/cloud-volume) packages. The hemibrainr package (https://github.com/flyconnectome/hemibrainr) was used to fetch each putMIP LN’s metadata and calculate each neuron’s dendrite-axon segregation index and flow centrality⁶⁸ using the recommended arguments.

Putative MIPergic LN connectomic analyses — intraglomerular input:output ratio analysis

Glomerular meshes based on PN dendrites were used for all subsequent analyses (input:output ratio by glomerulus, connectivity demographics, etc.)²⁸. To establish a input:output ratio for each glomerulus a given putMIP LN innervates, we extracted the number of input and output connections each putMIP LN has within each glomerulus by subsetting the connectors read in from the neuPrint database via the neurprintr “neuprint_read_neurons” function. These connectors were then filtered for their presence inside each glomerulus’ mesh XYZ coordinate space, segregated based on connection type (e.g., output), then finally summed. By analyzing the data in this manner, as opposed to simply considering the number of putMIP LN axons/dendrites within a given glomerulus, this analysis likely more closely captures putMIP LN input-vs.-output across the AL as Drosophila synapses are generally polyadic (reviewed in⁶²). To establish a given putMIP LN’s input:output ratio across all glomeruli, we used the following formula: (# of input connections - # of output connections)/(# of input connections + # of output connections). Therefore, values from −1 to 0 indicate the given putMIP LN sends more output within the given glomerulus. Conversely, values from 0 to 1 indicates the given putMIP LN receives more input within the glomerulus.

Putative MIPergic LN connectomic analyses — general upstream and downstream demographics analyses

To identify and compare the demographics of each putMIP LN’s upstream and downstream partners, putMIP LN connectivity data were first extracted using the hemibrainr “simple connectivity” function. The demographic of each presynaptic and postsynaptic partner was generally assigned according to the neuron’s accompanying name or type as listed on neuPrint, or by previously established cell type assignments²⁸. In cases where a neuron’s name or type was unannotated (i.e., NA), the neuron would be categorized as Unknown. The percentage of overall input a given putMIP LN receives from a given neuron category is the sum of connections from a given neuron category to the given putMIP LN divided by the total amount of input that given putMIP LN receives from all categories multiplied by 100%. Similar methods were applied for determining the percentage of overall output a given neuron category receives from a given putMIP LN.

Putative MIPergic LN connectomic analyses — putMIP LN input polarity analysis

To determine the amount of excitatory, inhibitory, and modulatory input a given putMIP LN receives within each glomerulus, we first categorized each presynaptic neuron as either excitatory, inhibitory, or modulatory based on the presynaptic neuron’s neuPrint name/type, previous immunohistochemistry results^{31,32,35,143,144,145,146}, and/or the category assigned in previous reports²⁸. However, we acknowledge several caveats to this analysis, such as: (1) this analysis does not account for co-transmission; (2) several glomeruli are truncated within the hemibrain AL²⁸; (3) although we consider all LNs as inhibitory as most are either GABAergic or glutamatergic (combined, these represent ~170/200 AL LNs)^{31,32,35,111,143,145}, there are ~4 tyrosine hydroxylase-immunoreactive (dopaminergic) and ~8–15 cholinergic and/or electrically coupled LNs in the AL^32,33,38,39; (4) although GABA can also act as an intrinsic modulator in the AL (reviewed by Lizbinski & Dacks¹⁴⁷), we only count GABAergic LNs as part of the “inhibitory input” category here; and, (5) we consider all ventral LNs analyzed here as being glutamatergic, but there are ~4 dopaminergic (tyrosine hydroxylase-immunoreactive) ventral LNs³². Once each presynaptic neuron’s chemical identity (excitatory, inhibitory, modulatory, or unknown) was determined, we used several approaches to assign these synapses to particular glomeruli. In the case of uniglomerular PNs (uPNs) and OSNs, we leveraged the single glomerulus innervation of these presynaptic neuron types to assign their synapse onto a given putMIP LN synapse to the presynaptic neuron’s home glomerulus. That is to say, OSN-to-putMIP LN and uPN-to-putMIP LN synapses were assigned to a glomerulus by: (1) using the home glomerulus assigned to a given presynaptic in the neuron’s neuPrint name/ type, or (2) by the home glomerulus assigned to the neuron in previous reports²⁸. For instance, if the presynaptic neuron was a cholinergic PN whose home glomerulus is DA2, and this DA2 PN synapses on a given putMIP LN five times, then those five synapses went to the overall excitatory input the given putMIP LN receives within DA2. Neurons were only excluded from this analysis if the presynaptic neuron’s home glomerulus was not previously identified²⁸. Once the polarity of the input type was established, we used the same methods as above for determining whether the XYZ coordinates of each putMIP LN’s synapse(s) with a given presynaptic partner were located in a given glomerulus. Synapse counts for each putMIP LN partner within the given glomerulus were then summed by type (excitatory, inhibitory, modulatory, or unknown), and the resulting total was divided by the total number of synapses the given putMIP LN makes within that glomerulus to establish percent excitatory, inhibitory input, or modulatory input.

SPR^MI13885-T2A-LexA::QFAD generation

The SPR^MI13885-T2A-LexA::QFAD fly line was established using previously described injections methods⁴¹. We also note that we also attempted to create an SPR-T2A-GAL4 using the pC-(lox2-attB2-SA-T2A-Gal4-Hsp70)3 construct (Addgene #62957), but no founders emerged (potentially owing to lethality when these construct elements are inserted in the SPR locus). Briefly, pBS-KS-attB2-SA(2)-T2A-LexA::QFAD-Hsp70 and ΦC31 helper plasmid DNA were co-injected into y¹, w^*, Mi{MIC}SPR^MI13885. pBS-KS-attB2-SA(2)-T2A-LexA::QFAD-Hsp70 (Addgene plasmid #62949) and pC-(lox2-attB2-SA-T2A-Gal4-Hsp70)3 (Addgene #62957) were gifts from Benjamin H. White (NIH). SPR^MI13885-T2A-LexA::QFAD transformants were isolated as depicted in Supplementary Fig. 7.

Single-cell RNA-sequencing (scRNA-seq) analysis of SPR expression

Single-cell transcriptomic data were accessed and downloaded from the SCope web interface (https://scope.aertslab.org) on 03/04/2022. Projection neuron clusters were re-identified as in each dataset’s original report^88,89,90. Transcript reads were exported log-transformed (log(1 + x)) and reads were counts-per-million (CPM) normalized. Projection neuron subpopulations were then identified within each scRNA-seq dataset using previously established marker genes^88,148,149.

in vivo calcium imaging — animal preparation

All calcium imaging experiments were performed on female flies ~1–5 days post-eclosion, and at room temperature. All physiology occurred within the animal’s ZT0 and ZT8. Animals of the proper genotype were collected and briefly anesthetized on ice. Once anesthetized, an animal was affixed to a custom-built holder with UV curable glue (BONDIC, M/N: SK8024). Our custom-built holder consists of a sheet of aluminum foil with a ~1 × 1 mm square (the imaging window) affixed to a 3D-printed design derived from similar designs described previously¹⁵⁰. Once mounted, a small window exposing the dorsal side of the brain was created, and covered with twice-filtered recording saline (in mM: 2 CaCl₂, 5 KCl, 5 HEPES, 8.2 MgCl₂, 108 NaCl, 4 NaHCO₃, 1 NaH₂PO₄, 10 sucrose, and 5 trehalose; adjusted pH: ~7.4)²⁹. After establishing the imaging window, the air sacs, fat bodies, and trachea covering the dorsal side of the brain - as well as Muscle 16 - were removed with fine forceps. With the exception of minimal epochs during the synthetic MIP bath application experiments (see below), the brain was continuously perfused with oxygenated (95%O₂/5%CO₂) recording saline using a Cole-Parmer Masterflex C/L (M/N: 77120-62) at a rate of ~2 mL min⁻¹.

in vivo calcium imaging — image acquisition

For one-photon imaging data (the majority of in vivo physiology data), data were acquired using a Prior Scientific Open Stand (M/N: H175) microscope mounted on Prior Scientific motorized translational stage (M/N: HZPKT1), and equipped with an Olympus 10x/0.30 UPlanFL N objective and an Olympus 60x/1.00 LUMPlanFL N water-immersion objective. A 470 nm CoolLED pE-100 (CoolLED Ltd., Hampshire, UK) was used as the light source. Each trial was captured with a Hamamatsu ORCA-4.0LT camera (Hamamatsu Phototonics, Hamamatsu, Japan), and consists of 40 1,024 × 1,024 frames acquired at a frame rate of ~9 Hz.

A portion of the R32F10-GAL4 odor panel experiments were also acquired using a custom-built two-photon system (Scientifica) equipped with a Mai Tai HP Ti:Sapphire laser (Spectra-Physics) and operated using ScanImage acquisition software (v.5.5; Vidrio Technologies). Emitted fluorescence was captured by a gallium arsenide phosphide (GaAsP) photomultiplier-tube detectors. Each trial consisted of 80 512 × 512 frames acquired at a frame rate of ~3.4 Hz. After data acquisition, a high-resolution z-stack (1,024 × 1,024) was acquired at ~0.21 Hz to enable post-hoc glomerulus identification as previously described^{29,40,151,152,153} (also, see below).

in vivo calcium imaging — odor preparation and delivery

All odor concentrations are reported as v/v dilutions in paraffin oil (J.T. Baker, VWR #JTS894), or autoclaved and twice-filtered distilled water (for diluting acids). For example, 10⁻² dilution indicates that one volume of an odor is diluted with 100 volumes of paraffin oil.

For one-photon imaging data (the majority of in vivo physiology data), dilutions were prepared in 2 mL odor vials (SUPELCO; P/N: 6020) that contained a final volume of 1 mL of diluted odor in paraffin oil every other day, or after two experiments (whichever came first). Odors were generally presented as previously described^75,77,119. Briefly, a carrier stream of carbon-filtered, dehumidified, air was presented at 2.2 L min⁻¹ to the fly continuously through an 8 mm Teflon tube placed ~1 cm away from the fly. A three-way solenoid (The Lee Company, P/N: LHDA1231315H) diverted a small portion of the airstream (0.2 L min⁻¹) through the headspace of an odor vial for 200 ms after triggering an external voltage command (TTL pulse) at frame 20 of the trial. Considering the above, the odor is diluted further (by 10-fold) prior to delivery to the animal. The odor stream joined the carrier stream 11 cm from the end of the tube, and the tube opening measured ~4 mm. Odor delivery during two-photon imaging was similar, but differed slightly in that: (1) odor cartridges (see below) instead of a 2 mL odor vial; (2) the continuous airstream was presented via a custom-built glass tube; and, (3) the TTL pulse occurred at frame 30 of the trial.

Methods for assessing preparation health and performing multiple odor trials conform to previous work^75,119. At the start of each experiment, the animal was presented a test odor (10⁻³ 2-heptanone) to assess the preparation’s health. Only the data collected from animals whose responses to this test odor were robust and did not dramatically change from baseline over the course of the experiment were used for further analysis. The only exceptions to this were those data collected in synthetic MIP bath application experiments (see below), since bath application of any modulator would likely result in network property changes that would consequently change olfactory responses. Therefore, the test odor was only initially presented to those animals used for synthetic peptide application experiments, so their initial olfactory response health could be assessed. Each experiment consisted of multiple odor trials (3 for OSNs; 4 for LNs) within a preparation which were then averaged to attain a within-animal response. These within-animal averages were subsequently averaged across many animals for subsequent statistical analysis, and “n” is reported as the number of animals. Each odor trial consisted of five 200 ms pulses of odor with a 1 ms interpulse interval. The same odor was never presented twice within 2 min to prevent depletion of the odor vial’s headspace. If multiple odors were to be tested, then they were presented randomly. If multiple concentrations of a given odor were to be tested, then the lower concentration was presented before the higher concentration. Air entered and exited each odor vial through a PEEK one-way check valve (The Lee Company, P/N: TKLA3201112H) connected to the vial by Teflon tubing. The odor delivery tube was flushed with clean air for 2 min when changing between odors/concentrations. As an additional preemptive measure, all odor delivery system components were hooked up to the house vacuum line overnight. The olfactometer used in two-photon data collection consisted of odor cartridges (a syringe housing a piece of filter paper that was doused in 10 µl of diluted odor) hooked into a custom glass carrier stream delivery tube as previously described¹⁵⁴.

in vivo calcium imaging — data analysis

All calcium imaging data were analyzed using a custom-made MATLAB script graciously provided by Marco Gallio (Northwestern University) and has been described previously^51,155,156. With the exception of any preparations that violated the aforementioned criteria (e.g., movement, diminishing prep health, etc.), no data points or outliers were excluded from our analyses. Generally, the number of flies to be used for experiments are not a limiting factor, therefore no statistical power analyses were used to pre-determine sample sizes. Regardless, our sample sizes are similar to those in previous reports that perform similar experiments^{30,51,157,158,159,160,161,162}. Before analyzing the data, a Gaussian low-pass filter (sigma=1), bleach correction (exponential fit), and image stabilizer algorithms were applied to the given trial’s raw ∆F/F signal. Similar preprocessing for two-photon microscopy data was similar, with the exception of a higher sigma during Gaussian low-pass filtering (sigma=2). A trial’s average fluorescence image was used as a guide to draw consistently sized circular regions-of-interest (ROI) within a given glomerulus. Calcium transients (∆F/F) within the ROI were measured as changes in fluorescence (∆F) normalized to baseline fluorescence (F, fluorescence intensity averaged across 2 sec just prior to odor onset). Within-animal responses were established by averaging across several odor trials in the given preparation (3 for OSNs; 4 for LNs). These within-animal responses were then pooled for each stimulus identity and concentration across animals. These pooled averages were used for all subsequent statistical analyses and the “n” is reported as the number of animals. Glomeruli were manually identified post-hoc by comparing acquired images to well-defined three-dimensional maps of the AL^163,164. Only the glomeruli that were reasonably identifiable were considered for analysis.

Myoinhibitory peptide (MIP) application experiments

MIP (MIP; EPTWNNLKGMW-amide) was custom made by GenScript (Piscataway, NJ, USA) at the highest purity available (>75%). The sequence we chose to use for MIP is identical to the sequence previous investigations have used when discerning the role of MIP in the Drosophila circadian system¹⁵⁸. In pilot experiments, we tested another sequence of MIP (RQAQGWNKFRGAW-amide) that was previously detected at the highest abundance by direct profiling of single ALs using mass spectrometry^26,165. Experimental results produced using peptide of either sequence were not qualitatively different, but all results reported here use the MIP previously used in circadian studies¹⁵⁸. To test how MIP application adjusts odor-evoked responses, a 1,000 µM working solution was made by diluting a small portion of the lyophilized peptide in nuclease-free water (ThermoScientific, #R0581). After testing the initial odor-evoked responses of the neurons being tested for a given experiment, the perfusion system was momentarily switched off so a small portion of our MIP working solution could be pressure injected into the AL to a final concentration of 10 µM. This final concentration was chosen for several reasons, which include: (1) we wished to remain consistent with other studies of peptidergic modulation in the Drosophila AL^29,30,157; (2) we wished to be consistent with studies on the effects of MIP in other circuits¹⁵⁸; and, (3) previous reports have already determined that our chosen effective concentration (10 µM) is the optimal concentration for testing the effect of MIP on Drosophila neurons¹⁵⁸. Ten minutes after MIP pressure injection, the animal’s odor-evoked responses were tested as before MIP injection, and then the perfusion system was switched back on. Ten minutes after turning the perfusion system back on, the animal’s odor-evoked responses were once again tested as they were initially. Re-testing the animal’s response to the test odor (10⁻³ 2-heptanone) at the end of these experiments could not be used as a reliable means for assessing prep health due to changes in circuit member responses induced by modulator bath application. Therefore, for these experiments no animal was tested for longer than the average time that animals were reliably healthy in the MIPergic LN odor panel experiments (~90 min). Furthermore, we believe these preparations remain healthy throughout the entire experimental epoch as ACV responses increase or do not significantly diminish over the course of the experimental epoch in many glomeruli (Fig. 1). It is notable that subtle differences exist between the”pre-MIP”/baseline peb-GAL4 GCaMP responses (Fig. 1) and peb-GAL4 > SPR-RNAi GCaMP responses (Fig. 8). The difference in baseline across these figures likely stems from: (1) subtle differences in the imaging plane which is inherent to these types of in vivo recordings across these experiments; and, (2) the difference in genotype of the animals being recorded in each respective figure. To address this, the within-figure baseline values are only ever internally compared by having within-animal measures across treatments (i.e., pre-MIP, MIP, and post-MIP). In doing so, the impact of MIP is with relation to the values from the same animal.

Quantification and statistical analyses

General approach

Statistical analyses were performed using R (v.4.1.3) in R Studio (v.2022.07.2). Values to be analyzed were concatenated in Excel before importing into the relevant analysis software. Statistical results are reported within the main text and/or figure legends. All statistical tests were two-tailed. All boxplots display the minimum, 25th-percentile, median, 75th-percentile, and maximum of the given data. Additional analysis details are provided for each set of experiments above. Where possible, values are given as mean ± SEM. Statistical significance is defined as: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Statistical analyses related to neuroanatomical experiments

The ComplexHeatmap package was used to hierarchically cluster glomeruli and individual MIPergic LN clones using Ward’s criteria and Euclidian distance. The ClustVis package (https://github.com/taunometsalu/ClustVis)¹¹⁷ was used to perform PCA on individual MIPergic LN innervation patterns. The “cor” function in the base-R stats package and the “rcorr” function in the Hmisc package were used to calculate statistically significant Pearson’s correlation coefficients for MIPergic LN pairwise glomerular innervation patterns. The ggpubr package’s “ggscatter” function was used to determine Pearson’s correlation coefficients and p-values when assessing correlations between: (1) effector/anti-MIP and MIPergic LN mCD8::GFP voxel density across all glomeruli, and (2) MIPergic LN glomerular innervation frequency as a function of each glomerulus’ volume. Adjusted R² values were calculated using the base-R stats package and correspond to how well each data being assessed for the given correlation analysis fit a linear model. The Shapiro-Wilk test (the rstatix package’s “shapiro_test” function) was used to evaluate any deviations from a normal distribution. Welch’s unpaired t-test was used to determine if MIPergic LNs preferentially innervate glomeruli based on inferred hedonic valence. A Kruskal-Wallis rank sum test followed by pairwise Bonferroni’s-corrected Dunn’s multiple comparisons test was used to determine if: (1) MIPergic LNs preferentially innervate based on the functional group found along the odorant that activates the given glomerulus’ odorant receptor; (2) SPR-GAL4::VP16 expression in antennae and maxillary palps significantly differs between males, mated females, and virgin females; (3) SPR-GAL4::VP16 expression in glutamatergic LNs between males, mated females, and virgin females; (4) SPR-T2A-GAL4 expression in maxillary palps significantly differs between males, mated females, and virgin females; and, (5) the number of MIPergic LNs differ between males, mated females, and virgin females. Welch’s one-way ANOVA with a Bonferroni multiple comparisons correction was used to assess statistically significant differences in SPR-T2A-GAL4 expression in antennae between males, mated females, and virgin females. A two-way ANOVA with a Greenhouse-Geisser sphericity correction followed by a Bonferroni’s multiple comparisons test was used to assess sexual dimorphism in MIPergic LN syt.eGFP or DenMark puncta density across glomeruli.

Statistical analyses related to physiology experiments

Background-subtracted changes in fluorescence over time (∆F/F) analyses were carried out using custom MATLAB scripts previously described^51,155, and are represented as individual traces overlaid by the mean with dilutant-only (e.g., paraffin oil-only) responses subtracted. Peak response (Fig. 3e) refers to the maximal ∆F/F value within the time of odor onset to ~1 s post-odor onset averaged across all animals. Area under the ∆F/F curve (AUC) was modified from previous reports¹⁶⁶, such that AUC was calculated using Simpson’s rule (“sintegral” function in the Bolstad2 package) as the integral of the ∆F/F traces from the beginning until 1 s after odor delivery with a baseline of 1 s before stimulus onset. To assess OSN odor-evoked response differences across MIP treatments, we first determined if normality could be assumed (as above). If normality could be assumed, then an omnibus repeated measures one-way ANOVA with a Greenhouse-Geisser sphericity correction was performed (RM one-way ANOVA) (“anova_test” function in rstatix). If significant differences were detected with the omnibus, then pairwise repeated measures t-tests (RM t-tests) with a Holm multiple comparisons correction were performed to identify which groups were statistically different. If normality could not be assumed, then a Friedman rank sum test followed by Holm-corrected paired two-sided Wilcoxon signed-rank test was performed. All effect sizes reported were calculated using either the “cohens_d” (for parametric data) or “friedman_effsize” (for non-parametric data) function from the rstatix package, which compute Cohen’s d or Kendall’s W, respectively.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Connectomic and scRNA-seq source data are available on neuPrint [https://neuprint.janelia.org/] and SCope [https://scope.aertslab.org/], respectively. All original light microscopy confocal scans used to create the figure panels are available at https://zenodo.org/10.5281/zenodo.8127341. Source data are provided with this paper.

Code availability

With the exception of code that was graciously provided to us by others, all code that was used to analyze or plot data is available at https://github.com/tsizemo2/Sizemoreetal2023.

References

Taghert, P. H. & Nitabach, M. N. Peptide Neuromodulation in Invertebrate Model Systems. Neuron 76, 82–97 (2012).
Article CAS PubMed Central PubMed Google Scholar
Katsukura, Y., Ando, H., David, C. N., Grimmelikhuijzen, C. J. P. & Sugiyama, T. Control of planula migration by LWamide and RFamide neuropeptides in Hydractinia echinata. J. Exp. Biol. 207, 1803–1810 (2004).
Article CAS PubMed Google Scholar
Golubovic, A. et al. A peptide-gated ion channel from the freshwater polyp Hydra. J. Biol. Chem. 282, 35098–35103 (2007).
Article CAS PubMed Google Scholar
Kass-Simon, G. & Pierobon, P. Cnidarian chemical neurotransmission, an updated overview. Comp. Biochem. Physiol. - A Mol. Integr. Physiol. 146, 9–25 (2007).
Article CAS PubMed Google Scholar
Watanabe, H., Fujisawa, T. & Holstein, T. W. Cnidarians and the evolutionary origin of the nervous system. Dev. Growth Differ. 51, 167–183 (2009).
Article CAS PubMed Google Scholar
Nassel, D., Pauls, D. & Huetteroth, W. Neuropeptides in modulation of Drosophila behavior: how to get a grip on their pleiotropic actions. Curr. Opin. Insect Sci. 36, 1–8 (2019).
Article PubMed Google Scholar
Conzelmann, M. et al. Neuropeptides regulate swimming depth of Platynereis larvae. Proc. Natl Acad. Sci. USA 108, E1174–E1183 (2011).
Zieger, E., Robert, N. S. M., Calcino, A. & Wanninger, A. Ancestral role of ecdysis-related neuropeptides in animal life cycle transitions. Curr. Biol. 31, 1–7 (2021).
Article Google Scholar
Moroz, L. L., Romanova, D. Y. & Kohn, A. B. Neural versus alternative integrative systems: molecular insights into origins of neurotransmitters. Philos. Trans. R. Soc. B. 376, 20190762 (2021).
Katz, P. S. & Lillvis, J. L. Reconciling the deep homology of neuromodulation with the evolution of behavior. Curr. Opin. Neurobiol. 29, 39–47 (2014).
Article CAS PubMed Google Scholar
De Bono, M. & Bargmann, C. I. Natural variation in a neuropeptide Y receptor homolog modifies social behavior and food response in C. elegans. Cell 94, 679–689 (1998).
Article PubMed Google Scholar
Inui, A. Feeding and body-weight regulation by hypothalamic neuropeptides—mediation of the actions of leptin. Trends Neurosci. 22, 62–67 (1999).
Article CAS PubMed Google Scholar
Wu, Q., Zhao, Z. & Shen, P. Regulation of aversion to noxious food by Drosophila neuropeptide Y- and insulin-like systems. Nat. Neurosci. 8, 1350–1355 (2005).
Article CAS PubMed Google Scholar
Jing, J. et al. From hunger to satiety: reconfiguration of a feeding network by Aplysia neuropeptide Y. J. Neurosci. 27, 3490–3502 (2007).
Article CAS PubMed Central PubMed Google Scholar
van den Pol, A. N. Neuropeptide transmission in brain circuits. Neuron 76, 98–115 (2012).
Article PubMed Central PubMed Google Scholar
Maeda, T. et al. Suppressive effects of dRYamides on feeding behavior of the blowfly, Phormia regina. Zool. Lett. 1, 1–10 (2015).
Article Google Scholar
Ohno, H. et al. Luqin-like RYamide peptides regulate food-evoked responses in C. Elegans. Elife 6, 1–23 (2017).
Article Google Scholar
Duvall, L. B., Ramos-Espiritu, L., Barsoum, K. E., Glickman, J. F. & Vosshall, L. B. Small-molecule agonists of Ae. aegypti neuropeptide Y receptor block mosquito biting. Cell 176, 687–701.e5 (2019).
Article CAS PubMed Central PubMed Google Scholar
Yokobori, E. et al. Neuropeptide Y Stimulates Food Intake in the Zebrafish, Danio rerio. J. Neuroendocrinol. 24, 766–773 (2012).
Article CAS PubMed Google Scholar
Nässel, D. R. & Zandawala, M. Endocrine cybernetics: neuropeptides as molecular switches in behavioural decisions. Open Biol. 12, 220174 (2022).
Nässel, D. R. & Zandawala, M. Recent advances in neuropeptide signaling in Drosophila, from genes to physiology and behavior. Prog. Neurobiol. 179, 101607 (2019).
Article PubMed Google Scholar
Baraban, S. C. & Tallent, M. K. Interneuron diversity series: interneuronal neuropeptides—endogenous regulators of neuronal excitability. Trends Neurosci. 27, 135–142 (2004).
Article CAS PubMed Google Scholar
Min, S. et al. Identification of a peptidergic pathway critical to satiety responses in drosophila. Curr. Biol. 26, 814–820 (2016).
Article CAS PubMed Google Scholar
Hussain, A., Üçpunar, H. K., Zhang, M., Loschek, L. F. & Grunwald Kadow, I. C. Neuropeptides modulate female chemosensory processing upon mating in Drosophila. PLoS Biol. 14, e1002455 (2016).
Article PubMed Central PubMed Google Scholar
Semmelhack, J. L. & Wang, J. W. Select Drosophila glomeruli mediate innate olfactory attraction and aversion. Nature 459, 218–223 (2009).
Article ADS CAS PubMed Central PubMed Google Scholar
Carlsson, M. A., Diesner, M., Schachtner, J. & Nässel, D. R. Multiple neuropeptides in the Drosophila antennal lobe suggest complex modulatory circuits. J. Comp. Neurol. 518, 3359–3380 (2010).
Article CAS PubMed Google Scholar
Wilson, R. I. & Laurent, G. Role of GABAergic inhibition in shaping odor-evoked spatiotemporal patterns in the Drosophila antennal lobe. J. Neurosci. 25, 9069–9079 (2005).
Article CAS PubMed Central PubMed Google Scholar
Schlegel, P. et al. Information flow, cell types and stereotypy in a full olfactory connectome. Elife https://doi.org/10.1101/2020.12.15.401257 (2021).
Root, C. M. et al. A presynaptic gain control mechanism fine-tunes olfactory behavior. Neuron 59, 311–321 (2008).
Article CAS PubMed Central PubMed Google Scholar
Ignell, R. et al. Presynaptic peptidergic modulation of olfactory receptor neurons in Drosophila. Proc. Natl Acad. Sci. USA 106, 13070–13075 (2009).
Article ADS CAS PubMed Central PubMed Google Scholar
Okada, R., Awasaki, T. & Ito, K. Gamma-aminobutyric acid (GABA)-mediated neural connections in the Drosophila antennal lobe. J. Comp. Neurol. 514, 74–91 (2009).
Article CAS PubMed Google Scholar
Chou, Y.-H. et al. Diversity and wiring variability of olfactory local interneurons in the Drosophila antennal lobe. Nat. Neurosci. 13, 439–449 (2010).
Article CAS PubMed Central PubMed Google Scholar
Yaksi, E. & Wilson, R. I. Electrical coupling between olfactory glomeruli. Neuron 67, 1034–1047 (2010).
Article CAS PubMed Central PubMed Google Scholar
Seki, Y., Rybak, J., Wicher, D., Sachse, S. & Hansson, B. S. Physiological and morphological characterization of local interneurons in the Drosophila antennal lobe. J. Neurophysiol. 104, 1007–1019 (2010).
Article PubMed Google Scholar
Das, A. et al. Identification and analysis of a glutamatergic local interneuron lineage in the adult Drosophila olfactory system. Neural Syst. Circuits 1, 4 (2011).
Article PubMed Central PubMed Google Scholar
Tanaka, N. K., Endo, K. & Ito, K. Organization of antennal lobe-associated neurons in adult Drosophila melanogaster brain. J. Comp. Neurol. 520, 4067–4130 (2012).
Article PubMed Google Scholar
Olsen, S. R., Bhandawat, V. & Wilson, R. I. Excitatory Interactions between Olfactory Processing Channels in the Drosophila Antennal Lobe. Neuron 54, 89–103 (2007).
Article CAS PubMed Central PubMed Google Scholar
Shang, Y., Claridge-Chang, A., Sjulson, L., Pypaert, M. & Miesenböck, G. Excitatory Local Circuits and Their Implications for Olfactory Processing in the Fly Antennal Lobe. Cell 128, 601–612 (2007).
Article CAS PubMed Central PubMed Google Scholar
Huang, J., Zhang, W., Qiao, W., Hu, A. & Wang, Z. Functional connectivity and selective odor responses of excitatory local interneurons in drosophila antennal lobe. Neuron 67, 1021–1033 (2010).
Article CAS PubMed Google Scholar
Root, C. M., Semmelhack, J. L., Wong, A. M., Flores, J. & Wang, J. W. Propagation of olfactory information in Drosophila. Proc. Natl Acad. Sci. USA 104, 11826–11831 (2007).
Article ADS CAS PubMed Central PubMed Google Scholar
Diao, F. et al. Plug-and-Play Genetic Access to Drosophila Cell Types using Exchangeable Exon Cassettes. Cell Rep. 10, 1410–1421 (2015).
Article CAS PubMed Central PubMed Google Scholar
Menuz, K., Larter, N. K., Park, J. & Carlson, J. R. An RNA-Seq Screen of the Drosophila Antenna Identifies a Transporter Necessary for Ammonia Detection. PLoS Genet. 10, 1410–1421 (2014).
Mohapatra, P. & Menuz, K. Molecular profiling of the Drosophila antenna reveals conserved genes underlying olfaction in insects. G3: Genes, Genomes, Genet. 9, 3753–3771 (2019).
Article CAS Google Scholar
McLaughlin, C. N. et al. Single-cell transcriptomes of developing and adult olfactory receptor neurons in drosophila. Elife 10, 1–37 (2021).
Article Google Scholar
Gupta, A., Wang, Y. & Markram, H. Organizing principles for a diversity of GABAergic interneurons and synapses in the neocortex. Science 287, 273–278 (2000).
Article ADS CAS PubMed Google Scholar
Markram, H. et al. Interneurons of the neocortical inhibitory system. Nat. Rev. Neurosci. 5, 793–807 (2004).
Article CAS PubMed Google Scholar
Jenett, A. et al. A GAL4-Driver Line Resource for Drosophila Neurobiology. Cell Rep. 2, 991–1001 (2012).
Article CAS PubMed Central PubMed Google Scholar
Rao, S., Lang, C., Levitan, E. S. & Deitcher, D. L. Visualization of neuropeptide expression, transport, and exocytosis in Drosophila melanogaster. J. Neurobiol. 49, 159–172 (2001).
Article CAS PubMed Google Scholar
Nern, A., Pfeiffer, B. D. & Rubin, G. M. Optimized tools for multicolor stochastic labeling reveal diverse stereotyped cell arrangements in the fly visual system. Proc. Natl Acad. Sci. USA 112, E2967–E2976 (2015).
Article ADS CAS PubMed Central PubMed Google Scholar
Gallio, M., Ofstad, T. A., Macpherson, L. J., Wang, J. W. & Zuker, C. S. The coding of temperature in the Drosophila brain. Cell 144, 614–624 (2011).
Article CAS PubMed Central PubMed Google Scholar
Frank, D. D. et al. Early Integration of Temperature and Humidity Stimuli in the Drosophila Brain. Curr. Biol. 27, 2381–2388.e4 (2017).
Article CAS PubMed Central PubMed Google Scholar
Marin, E. C. et al. Connectomics Analysis Reveals First-, Second-, and Third-Order Thermosensory and Hygrosensory Neurons in the Adult Drosophila Brain. Curr. Biol. 30, 3167–3182.e4 (2020).
Article CAS PubMed Central PubMed Google Scholar
Hallem, E. A. & Carlson, J. R. Coding of Odors by a Receptor Repertoire. Cell 125, 143–160 (2006).
Article CAS PubMed Google Scholar
Silbering, A. F., Okada, R., Ito, K. & Galizia, C. G. Olfactory information processing in the Drosophila antennal lobe: anything goes? J. Neurosci. 28, 13075–13087 (2008).
Article CAS PubMed Central PubMed Google Scholar
Haddad, R. et al. Global features of neural activity in the olfactory system form a parallel code that predicts olfactory behavior and perception. J. Neurosci. 30, 9017–9026 (2010).
Article CAS PubMed Central PubMed Google Scholar
Nicolaï, L. J. J. et al. Genetically encoded dendritic marker sheds light on neuronal connectivity in Drosophila. Proc. Natl Acad. Sci. USA 107, 20553–20558 (2010).
Article ADS PubMed Central PubMed Google Scholar
Zhang, Y. Q., Rodesch, C. K. & Broadie, K. Living synaptic vesicle marker: Synaptotagmin-GFP. Genesis 34, 142–145 (2002).
Article CAS PubMed Google Scholar
Schlegel, P., Costa, M. & Jefferis, G. S. X. E. Learning from connectomics on the fly. Curr. Opin. Insect Sci. 24, 96–105 (2017).
Article PubMed Google Scholar
Meinertzhagen, I. A. Of what use is connectomics? A personal perspective on the Drosophila connectome. J. Exp. Biol. 221, jeb164954 (2018).
Clements, J. et al. neuPrint: Analysis Tools for EM Connectomics. bioRxiv 1–20 https://doi.org/10.1101/2020.01.16.909465 (2020).
Scheffer, L. et al. A connectome and analysis of the adult drosophila central brain. Elife 1–83 https://doi.org/10.1101/2020.04.07.030213 (2020).
Prokop, A. & Meinertzhagen, I. A. Development and structure of synaptic contacts in Drosophila. Semin Cell Dev. Biol. 17, 20–30 (2006).
Article CAS PubMed Google Scholar
Merighi, A. Costorage of high molecular weight neurotransmitters in large dense core vesicles of mammalian neurons. Front Cell Neurosci. 12, 1–7 (2018).
Article Google Scholar
Flanagan, D. & Mercer, A. R. Morphology and response characteristics of neurones in the deutocerebrum of the brain in the honeybee Apis mellifera. J. Comp. Physiol. A 164, 483–494 (1989).
Article Google Scholar
Fonta, C., Sun, X. J. & Masson, C. Morphology and spatial distribution of bee antennal lobe interneurones responsive to odours. Chem. Senses 18, 101–119 (1993).
Article Google Scholar
Sachse, S. & Galizia, C. G. Role of inhibition for temporal and spatial odor representation in olfactory output neurons: A calcium imaging study. J. Neurophysiol. 87, 1106–1117 (2002).
Article PubMed Google Scholar
Galizia, C. G. & Kimmerle, B. Physiological and morphological characterization of honeybee olfactory neurons combining electrophysiology, calcium imaging and confocal microscopy. J. Comp. Physiol. A Neuroethol. Sens Neural Behav. Physiol. 190, 21–38 (2004).
Article CAS PubMed Google Scholar
Schneider-Mizell, C. M. et al. Quantitative neuroanatomy for connectomics in Drosophila. Elife 5, 1–36 (2016).
Article Google Scholar
Ding, H., Smith, R. G., Poleg-Polsky, A., Diamond, J. S. & Briggman, K. L. Species-specific wiring for direction selectivity in the mammalian retina. Nature 535, 105–110 (2016).
Article ADS CAS PubMed Central PubMed Google Scholar
Frechter, S. et al. Functional and anatomical specificity in a higher olfactory centre. Elife 8, 1–39 (2019).
Article Google Scholar
Barnes, C. L., Bonnéry, D. & Cardona, A. Synaptic counts approximate synaptic contact area in Drosophila. PLoS One 17, e0266064 (2022).
Article CAS PubMed Central PubMed Google Scholar
Lyu, C., Abbott, L. F. & Maimon, G. Building an allocentric travelling direction signal via vector computation. Nature 601, 92–97 (2022).
Article ADS CAS PubMed Google Scholar
Holler, S., Köstinger, G., Martin, K. A. C., Schuhknecht, G. F. P. & Stratford, K. J. Structure and function of a neocortical synapse. Nature 591, 111–116 (2021).
Article ADS CAS PubMed Google Scholar
Liu, T. X., Davoudian, P. A., Lizbinski, K. M. & Jeanne, J. M. Connectomic features underlying diverse synaptic connection strengths and subcellular computation. Curr. Biol. 32, 559–569.e5 (2022).
Article CAS PubMed Google Scholar
Hong, E. J. & Wilson, R. I. Simultaneous encoding of odors by channels with diverse sensitivity to inhibition. Neuron 85, 573–589 (2015).
Article CAS PubMed Central PubMed Google Scholar
Wilson, R. I., Turner, G. C. & Laurent, G. Transformation of Olfactory Representations in the Drosophila Antennal Lobe. Science 303, 366–371 (2004).
Article ADS CAS PubMed Google Scholar
Bhandawat, V., Olsen, S. R., Gouwens, N. W., Schlief, M. L. & Wilson, R. I. Sensory processing in the Drosophila antennal lobe increases reliability and separability of ensemble odor representations. Nat. Neurosci. 10, 1474–1482 (2007).
Article CAS PubMed Central PubMed Google Scholar
Seki, Y. et al. Olfactory coding from the periphery to higher brain centers in the Drosophila brain. BMC Biol. 15, 56 (2017).
Article PubMed Central PubMed Google Scholar
Yapici, N., Zimmer, M. & Domingos, A. I. Cellular and molecular basis of decision-making. EMBO Rep. 10, 1023–1035 (2014).
Article Google Scholar
Yang, C. H. et al. Control of the Postmating Behavioral Switch in Drosophila Females by Internal Sensory Neurons. Neuron 61, 519–526 (2009).
Article CAS PubMed Central PubMed Google Scholar
Kim, Y.-J. Y.-C. et al. MIPs are ancestral ligands for the sex peptide receptor. Proc. Natl Acad. Sci. 107, 6520–6525 (2010).
Article ADS CAS PubMed Central PubMed Google Scholar
Poels, J. et al. Myoinhibiting peptides are the ancestral ligands of the promiscuous Drosophila sex peptide receptor. Cell. Mol. Life Sci. 67, 3511–3522 (2010).
Article CAS PubMed Google Scholar
Katow, H., Takahashi, T., Saito, K. & Tanimoto, H. Tango knock-ins visualize endogenous activity of G protein-coupled receptors in Drosophila Tango knock-ins visualize endogenous activity of G protein-coupled receptors. J. Neurogenet. 0, 1–8 (2019).
Google Scholar
Joseph, R. M. & Carlson, J. R. Drosophila Chemoreceptors: A Molecular Interface Between the Chemical World and the Brain. Trends Genet. 31, 683–695 (2015).
Article CAS PubMed Central PubMed Google Scholar
Schmidt, H. R. & Benton, R. Molecular mechanisms of olfactory detection in insects: Beyond receptors: Insect olfactory detection mechanisms. Open Biol. 10, 200252 (2020).
Grabe, V. et al. Elucidating the Neuronal Architecture of Olfactory Glomeruli in the Drosophila Antennal Lobe. Cell Rep. 16, 3401–3413 (2016).
Article CAS PubMed Google Scholar
Ameku, T. et al. Midgut-derived neuropeptide F controls germline stem cell proliferation in a mating-dependent manner. PLoS Biol. 16, e2005004 (2018).
Article PubMed Central PubMed Google Scholar
Li, H. et al. Classifying Drosophila Olfactory Projection Neuron Resource Classifying Drosophila Olfactory Projection Neuron Subtypes by Single-Cell RNA Sequencing. Cell 171, 1206–1207.e22 (2017).
Article CAS PubMed Central PubMed Google Scholar
Croset, V., Treiber, C. D. & Waddell, S. Cellular diversity in the Drosophila midbrain revealed by single-cell transcriptomics. Elife 7, 1–31 (2018).
Article Google Scholar
Davie, K. et al. A Single-Cell Transcriptome Atlas of the Aging Drosophila Brain. Cell 982–998 https://doi.org/10.1016/j.cell.2018.05.057 (2018).
Yapici, N., Kim, Y.-J., Ribeiro, C. & Dickson, B. J. A receptor that mediates the post-mating switch in Drosophila reproductive behaviour. Nature 451, 33–37 (2008).
Article PubMed Google Scholar
Goodman, C. S. Isogenic grasshoppers: genetic variability in the morphology of identified neurons. J. Comp. Neurol. 182, 681–705 (1978).
Article CAS PubMed Google Scholar
Goeritz, M. L., Bowers, M. R., Slepian, B. & Marder, E. Neuropilar projections of the anterior gastric receptor neuron in the stomatogastric ganglion of the Jonah crab, Cancer borealis. PLoS One 8, 1–15 (2013).
Article Google Scholar
Otopalik, A. G., Sutton, A. C., Banghart, M. & Marder, E. When complex neuronal structures may not matter. Elife 6, 1–29 (2017).
Article Google Scholar
Otopalik, A. G. et al. Sloppy morphological tuning in identified neurons of the crustacean stomatogastric ganglion. Elife 6, 1–32 (2017).
Article Google Scholar
Marder, E., Goeritz, M. L. & Otopalik, A. G. Robust circuit rhythms in small circuits arise from variable circuit components and mechanisms. Curr. Opin. Neurobiol. 31, 156–163 (2015).
Article CAS PubMed Google Scholar
Otopalik, A. G., Pipkin, J. & Marder, E. Neuronal morphologies built for reliable physiology in a rhythmic motor circuit. Elife 8, 1–24 (2019).
Article Google Scholar
Thuma, J. B., White, W. E., Hobbs, K. H. & Hooper, S. L. Pyloric neuron morphology in the stomatogastric ganglion of the lobster, Panulirus interruptus. Brain Behav. Evol. 73, 26–42 (2009).
Article PubMed Google Scholar
Daur, N., Bryan, A. S., Garcia, V. J. & Bucher, D. Short-term synaptic plasticity compensates for variability in number of motor neurons at a neuromuscular junction. J. Neurosci. 32, 16007–16017 (2012).
Article CAS PubMed Central PubMed Google Scholar
Caron, S. J. C., Ruta, V., Abbott, L. F. & Axel, R. Random convergence of olfactory inputs in the Drosophila mushroom body. Nature 497, 113–117 (2013).
Article ADS CAS PubMed Central PubMed Google Scholar
Linneweber, G. A. et al. A neurodevelopmental origin of behavioral individuality in the Drosophila visual system. Science 367, 1112–1119 (2020).
Article ADS CAS PubMed Google Scholar
Ambros-Ingerson, J. & Holmes, W. R. Analysis and comparison of morphological reconstructions of hippocampal field CA1 pyramidal cells. Hippocampus 15, 302–315 (2005).
Article PubMed Google Scholar
Honegger, K. & de Bivort, B. Stochasticity, individuality and behavior. Curr. Biol. 28, R8–R12 (2018).
Article CAS PubMed Google Scholar
Honegger, K. S., Smith, M. A.-Y., Churgin, M. A., Turner, G. C. & de Bivort, B. L. Idiosyncratic neural coding and neuromodulation of olfactory individuality in Drosophila. Proc. Natl Acad. Sci. 117, 23292–23297 (2019).
Article ADS PubMed Central PubMed Google Scholar
Rihani, K. & Sachse, S. Shedding Light on Inter-Individual Variability of Olfactory Circuits in Drosophila. Front. Behav. Neurosci. 16, 1–11 (2022).
Stern, S., Kirst, C. & Bargmann, C. I. Neuromodulatory control of long-term behavioral patterns and individuality across development. Cell 171, 1649–1662.e10 (2017).
Article CAS PubMed Google Scholar
Pantoja, C. et al. Neuromodulatory Regulation of Behavioral Individuality in Zebrafish. Neuron 91, 587–601 (2016).
Article CAS PubMed Central PubMed Google Scholar
Olsen, S. R. & Wilson, R. I. Lateral presynaptic inhibition mediates gain control in an olfactory circuit. Nature 452, 956–960 (2008).
Article ADS CAS PubMed Central PubMed Google Scholar
Lima, S. Q. & Miesenböck, G. Remote control of behavior through genetically targeted photostimulation of neurons. Cell 121, 141–152 (2005).
Article CAS PubMed Google Scholar
Ding, K. et al. Imaging neuropeptide release at synapses with a genetically engineered reporter. Elife 8, 1–15 (2019).
CAS Google Scholar
Liu, W. W. & Wilson, R. I. Glutamate is an inhibitory neurotransmitter in the Drosophila olfactory system. Proc. Natl Acad. Sci. 110, 10294–10299 (2013).
Article ADS CAS PubMed Central PubMed Google Scholar
Nassel, D. R. Substrates for neuronal cotransmission with neuropeptides and small molecule neurotransmitters in drosophila. Front. Cell Neurosci. 12, 1–26 (2018).
Article Google Scholar
Marder, E. Neuromodulation of neuronal circuits: back to the future. Neuron 76, 1–11 (2012).
Article CAS PubMed Central PubMed Google Scholar
Nusbaum, M. P., Blitz, D. M. & Marder, E. Functional consequences of neuropeptide and small-molecule co-transmission. Nat. Rev. Neurosci. https://doi.org/10.1038/nrn.2017.56 (2017).
Kupfermann, I. Functional Studies of Cotransmission. Physiol. Rev. 71, 683–711 (1991).
Article CAS PubMed Google Scholar
Su, C. Y., Menuz, K. & Carlson, J. R. Olfactory perception: receptors, cells, and circuits. Cell 139, 45–59 (2009).
Article CAS PubMed Central PubMed Google Scholar
Bell, J. S. & Wilson, R. I. Behavior Reveals Selective Summation and Max Pooling among Olfactory Processing Channels. Neuron 91, 425–438 (2016).
Article CAS PubMed Central PubMed Google Scholar
Fişek, M. & Wilson, R. I. Stereotyped connectivity and computations in higher-order olfactory neurons. Nat. Neurosci. 17, 280–288 (2014).
Article PubMed Google Scholar
Jeanne, J. M., Fisek, M. & Wilson, R. I. The Organization of Projections from Olfactory Glomeruli onto Higher-Order Neurons. Neuron 96, 1198–1213 (2018).
Article Google Scholar
Liessem, S. et al. Neuromodulation Can Be Simple: Myoinhibitory Peptide, Contained in Dedicated Regulatory Pathways,Is the Only Neurally-Mediated Peptide Modulator of Stick Insect Leg Muscle. J. Neurosci. 41, 2911–2929 (2021).
Article CAS PubMed Central PubMed Google Scholar
Smith, S. J. et al. Single-cell transcriptomic evidence for dense intracortical neuropeptide networks. Elife 8, 1–35 (2019).
Article Google Scholar
Sizemore, T. R. & Dacks, A. M. Serotonergic Modulation Differentially Targets Distinct Network Elements within the Antennal Lobe of Drosophila melanogaster. Sci. Rep. 6, 37119 (2016).
Article ADS CAS PubMed Central PubMed Google Scholar
Peng, H., Ruan, Z., Long, F., Simpson, J. H. & Myers, E. W. V3D enables real-time 3D visualization and quantitative analysis of large-scale biological image data sets. Nat. Biotechnol. 28, 348–353 (2010).
Article CAS PubMed Central PubMed Google Scholar
Wan, Y. et al. FluoRender: joint freehand segmentation and visualization for many-channel fluorescence data analysis. BMC Bioinforma. 18, 1–15 (2017).
Article Google Scholar
Erdős, P. L. & Rényi, A. On a classical problem of probability theory. Mathematics Ser. A, 215–220 (1961).
MathSciNet MATH Google Scholar
Laissue, P. P. et al. Three-dimensional reconstruction of the antennal lobe in Drosophila melanogaster. J. Comp. Neurol. 405, 543–552 (1999).
Article CAS PubMed Google Scholar
Couto, A., Alenius, M. & Dickson, B. J. Molecular, Anatomical, and Functional Organization of the Drosophila Olfactory System. Curr. Biol. 15, 1535–1547 (2005).
Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849 (2016).
Article CAS PubMed Google Scholar
Bates, A. S. et al. Complete Connectomic Reconstruction of Olfactory Projection Neurons in the Fly Brain. Curr. Biol. 30, 3183–3199.e6 (2020).
Article CAS PubMed Central PubMed Google Scholar
Stensmyr, M. C. et al. A conserved dedicated olfactory circuit for detecting harmful microbes in drosophila. Cell 151, 1345–1357 (2012).
Article CAS PubMed Google Scholar
Min, S., Ai, M., Shin, S. A. & Suh, G. S. B. Dedicated olfactory neurons mediating attraction behavior to ammonia and amines in Drosophila. Proc. Natl Acad. Sci. 110, E1321–E1329 (2013).
Article ADS CAS PubMed Central PubMed Google Scholar
Ebrahim, S. A. M. et al. Drosophila Avoids Parasitoids by Sensing Their Semiochemicals via a Dedicated Olfactory Circuit. PLoS Biol. 13, 1–18 (2015).
Article Google Scholar
Mansourian, S. et al. Fecal-Derived Phenol Induces Egg-Laying Aversion in Drosophila. Curr. Biol. 26, 2762–2769 (2016).
Article CAS PubMed Google Scholar
Vulpe, A. et al. An ammonium transporter is a non-canonical olfactory receptor for ammonia. Curr. Biol. 1–9 https://doi.org/10.1016/j.cub.2021.05.025 (2021).
Knaden, M., Strutz, A., Ahsan, J., Sachse, S. & Hansson, B. S. Spatial Representation of Odorant Valence in an Insect Brain. Cell Rep. 1, 392–399 (2012).
Article CAS PubMed Google Scholar
Münch, D. & Galizia, C. G. DoOR 2.0 - Comprehensive Mapping of Drosophila melanogaster Odorant Responses. Sci. Rep. 6, 1–14 (2016).
Article Google Scholar
van Breugel, F., Huda, A. & Dickinson, M. H. Distinct activity-gated pathways mediate attraction and aversion to CO2 in Drosophila. Nature 564, 420–424 (2018).
Article ADS PubMed Central PubMed Google Scholar
Dawyuan Han, D., Stein, D. & Stevens, L. M. Investigating the function of follicular subpopulations during Drosophila oogenesis through hormone-dependent enhancer-targeted cell ablation. Development 127, 573–583 (2000).
Article PubMed Google Scholar
Bates, A. S. et al. The natverse, a versatile toolbox for combining and analysing neuroanatomical data. Elife 9, 1–35 (2020).
Article Google Scholar
Coates, K. E. et al. Identified Serotonergic Modulatory Neurons Have Heterogeneous Synaptic Connectivity within the Olfactory System of Drosophila. J. Neurosci. 37, 7318–7331 (2017).
Article CAS PubMed Central PubMed Google Scholar
Chiang, A. S. et al. Three-dimensional reconstruction of brain-wide wiring networks in drosophila at single-cell resolution. Curr. Biol. 21, 1–11 (2011).
Article MathSciNet CAS PubMed Google Scholar
Costa, M., Manton, J. D., Ostrovsky, A. D., Prohaska, S. & Jefferis, G. S. X. E. NBLAST: Rapid, Sensitive Comparison of Neuronal Structure and Construction of Neuron Family Databases. Neuron 91, 293–311 (2016).
Article CAS PubMed Central PubMed Google Scholar
Stocker, R. F., Heimbeck, G., Gendre, N. & de Belle, J. S. Neuroblast ablation in Drosophila P[GAL4] lines reveals origins of olfactory interneurons. J. Neurobiol. 32, 443–456 (1997).
Article CAS PubMed Google Scholar
Dacks, A. M., Christensen, T. A. & Hildebrand, J. G. Phylogeny of a Serotonin-Immunoreactive Neuron in the Primary Olfactory Center of the Insect Brain. J. Comp. Neurol. 498, 727–746 (2006).
Article CAS PubMed Google Scholar
Tanaka, N. K., Ito, K. & Stopfer, M. Odor-evoked neural oscillations in Drosophila are mediated by widely branching interneurons. J. Neurosci. 29, 8595–8603 (2009).
Article CAS PubMed Central PubMed Google Scholar
Sen, S., Biagini, S., Reichert, H. & VijayRaghavan, K. Orthodenticle is required for the development of olfactory projection neurons and local interneurons in Drosophila. Biol. Open 3, 711–717 (2014).
Article PubMed Central PubMed Google Scholar
Lizbinski, K. M. & Dacks, A. M. Intrinsic and Extrinsic Neuromodulation of Olfactory Processing. Front. Cell Neurosci. 11, 1–11 (2018).
Article Google Scholar
Komiyama, T., Johnson, W. A., Luo, L. & Jefferis, G. S. X. E. From lineage to wiring specificity: POU domain transcription factors control precise connections of Drosophila olfactory projection neurons. Cell 112, 157–167 (2003).
Article CAS PubMed Google Scholar
Komiyama, T. & Luo, L. Intrinsic Control of Precise Dendritic Targeting by an Ensemble of Transcription Factors. Curr. Biol. 17, 278–285 (2007).
Article CAS PubMed Google Scholar
Weir, P. T. et al. Anatomical reconstruction and functional imaging reveal an ordered array of skylight polarization detectors in drosophila. J. Neurosci. 36, 5397–5404 (2016).
Article CAS PubMed Central PubMed Google Scholar
Wang, J. W., Wong, A. M., Flores, J., Vosshall, L. B. & Axel, R. Two-photon calcium imaging reveals an odor-evoked map of activity in the fly brain. Cell 112, 271–282 (2003).
Article CAS PubMed Google Scholar
Dacks, A. M., Green, D. S., Root, C. M., Nighorn, A. J. & Wang, J. W. Serotonin modulates olfactory processing in the antennal lobe of Drosophila. J. Neurogenet. 23, 366–377 (2009).
Article CAS PubMed Central PubMed Google Scholar
Root, C. M., Ko, K. I., Jafari, A. & Wang, J. W. Presynaptic Facilitation by Neuropeptide Signaling Mediates Odor-Driven Food Search. Cell 145, 133–144 (2011).
Article CAS PubMed Central PubMed Google Scholar
Dacks, A. M., Christensen, T. A. & Hildebrand, J. G. Modulation of olfactory information processing in the antennal lobe of Manduca sexta by serotonin. J. Neurophysiol. 99, 2077–2085 (2008).
Article CAS PubMed Google Scholar
Frank, D. D., Jouandet, G. C., Kearney, P. J., MacPherson, L. J. & Gallio, M. Temperature representation in the Drosophila brain. Nature 519, 358–361 (2015).
Article ADS CAS PubMed Central PubMed Google Scholar
Alpert, M. H. et al. A Circuit Encoding Absolute Cold Temperature in Drosophila. Curr. Biol. 30, 2275–2288.e5 (2020).
Article CAS PubMed Central PubMed Google Scholar
Ko, K. I. et al. Starvation promotes concerted modulation of appetitive olfactory behavior via parallel neuromodulatory circuits. Elife 4, 1–17 (2015).
Article Google Scholar
Oh, Y. et al. A Homeostatic Sleep-Stabilizing Pathway in Drosophila Composed of the Sex Peptide Receptor and Its Ligand, the Myoinhibitory Peptide. PLoS Biol. 12, e1001974 (2014).
Badel, L., Ohta, K., Tsuchimoto, Y. & Kazama, H. Decoding of Context-Dependent Olfactory Behavior in Drosophila. Neuron 91, 155–167 (2016).
Article CAS PubMed Google Scholar
Díaz, M. M., Schlichting, M., Abruzzi, K. C., Long, X. & Rosbash, M. Allatostatin-C/AstC-R2 Is a Novel Pathway to Modulate the Circadian Activity Pattern in Drosophila. Curr. Biol. 29, 13–22.e3 (2019).
Article PubMed Google Scholar
Zandawala, M. et al. A neuroendocrine pathway modulating osmotic stress in Drosophila. PLoS Genetics 17, e1009425 (2021).
Lin, H. H. et al. A nutrient-specific gut hormone arbitrates between courtship and feeding. Nature 602, 632–638 (2022).
Article ADS CAS PubMed Central PubMed Google Scholar
Fishilevich, E. & Vosshall, L. B. Genetic and Functional Subdivision of the Drosophila Antennal Lobe Genetic and Functional Subdivision of the Drosophila Antennal Lobe. Curr. Biol. 15, 1548–1553 (2005).
Article CAS PubMed Google Scholar
Grabe, V., Strutz, A., Baschwitz, A., Hansson, B. S. & Sachse, S. Digital in vivo 3D atlas of the antennal lobe of Drosophila melanogaster. J. Comp. Neurol. 523, 530–544 (2015).
Article CAS PubMed Google Scholar
Yew, J. Y. et al. Analysis of neuropeptide expression and localization in adult Drosophila melanogaster central nervous system by affinity cell-capture mass spectrometry (Journal of Proteome Research (2009) 8, (1271-1284)). J. Proteome Res. 8, 3786 (2009).
Article CAS Google Scholar
Nojima, T. et al. A sex-specific switch between visual and olfactory inputs underlies adaptive sex differences in behavior. Curr. Biol. 31, 1175–1191.e6 (2021).
Article CAS PubMed Central PubMed Google Scholar

Download references

Acknowledgements

We are grateful to Kristyn Lizbinski, Eric Horstick, and John Carlson for the helpful comments given on earlier manuscript drafts. Flies were kindly shared or acquired from: Shu Kondo, David Krantz, Michael Texada, Jim Truman, Rachel Wilson, Quentin Gaudry, the Bloomington Drosophila Stock centers (NIH P40OD018537), and the Janelia Fly Light project. Christian Wegener kindly provided the MIP antibody that was developed by Manfred Eckert. Marco Gallio and Rachel Wilson graciously provided analysis scripts. Karen Menuz gave helpful insight in the interpretation of RNA-sequencing results from prior work. Stephen Plaza, Alexander Bates, Marta Costa, Greg Jefferis, and Philipp Schlegel gave helpful advice and instruction regarding connectomic analyses, and graciously shared odor scene and valence information. Our connectomic analyses also benefited from discussions in the nat-user community, and a workshop organized by the Virtual Fly Brain and Drosophila connectomics teams (virtualflybrain.org; WT 208379/Z/17/Z). Fengqiu Diao gave technical advice for creating the SPR^MI13885-T2A-LexA::QFAD Trojan exon transgenic animals. Kristyn Lizbinski and James Jeanne graciously provided advice and equipment specifications that lead to the construction and use of the olfactometer system described here. Kristyn Lizbinski, James Jeanne, Marco Gallio, Mehmet Keles, and Gaby Maimon provided invaluable technical advice for performing in vivo physiology. Kevin Daly kindly loaned equipment to us for our purposes and gave invaluable technical advice regarding their use. We would also like to thank Jacob Ralston for assisting by collecting confocal scans of AL anti-MIP labeling from R32F10-GAL4 diphtheria toxin ablation experiments. This work was supported by a Grant-In-Aid of Research (G20141015669888) from Sigma Xi, The Scientific Research Society (T.R.S.), start-up funds from WVU (A.M.D.), a National Institutes of Health R03 DC013997 (A.M.D.), a National Institutes of Health R01 DC016293 (A.M.D.), and two AFOSR DURIP awards (FA9550-19-1-0179 and FA9550-20-1-0098) (A.M.D.).

Author information

Tyler R. Sizemore
Present address: Department of Molecular, Cellular, and Developmental Biology, Yale Science Building, Yale University, New Haven, CT, 06520-8103, USA

Authors and Affiliations

Department of Biology, Life Sciences Building, West Virginia University, Morgantown, WV, 26506, USA
Tyler R. Sizemore, Julius Jonaitis & Andrew M. Dacks
Department of Neuroscience, West Virginia University, Morgantown, WV, 26506, USA
Andrew M. Dacks

Authors

Tyler R. Sizemore
View author publications
You can also search for this author in PubMed Google Scholar
Julius Jonaitis
View author publications
You can also search for this author in PubMed Google Scholar
Andrew M. Dacks
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

T.R.S. conceived and implemented all experiments, analyzed the subsequent data, prepared figures, and wrote and edited the manuscript. J.J. acquired data that contributed to Fig. 5. A.M.D. conceived experiments, acquired funding, and helped with manuscript edits.

Corresponding authors

Correspondence to Tyler R. Sizemore or Andrew M. Dacks.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Sizemore, T.R., Jonaitis, J. & Dacks, A.M. Heterogeneous receptor expression underlies non-uniform peptidergic modulation of olfaction in Drosophila. Nat Commun 14, 5280 (2023). https://doi.org/10.1038/s41467-023-41012-3

Download citation

Received: 02 February 2023
Accepted: 21 August 2023
Published: 30 August 2023
DOI: https://doi.org/10.1038/s41467-023-41012-3

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.